197105
197105 197105
197105 The Eye | File Listing
User Manual: 197105
Open the PDF directly: View PDF 
.
Page Count: 52
| Download | |
| Open PDF In Browser | View PDF |
<;1 /' /
SCIENCE & TECHNOLOGY
May, 1971
Vol. 20, No .5
co
IN THIS ISSUE:
The I nformation Revolution
- And the Bill of Rights
Programming and Mini-Computer
Costs
Cost Savings Possible in Data
Preparation
Maybe the Computers Can Save Us
After All
Computer-Controller Interface
PE RIODICALS SEC 1263399045
3 2 21SkRa:
0104
180 W SAN CARLOS Sf
*01271
SAN JOSF CA
95113
5" ;
Nf
01
in
m
te
al
Peter Denning, 29, is an Assistant
Professor of Electrical Engineering at
Princeton. He's also an ACM
member and chairman of our
committee on special interest groups
and committees (SIGs/SICs). He
wasn't always as active in ACM.
"I joined in 1965 while working on my
thesis," says Peter. "Mainly for
technical material and a chance to
meet other computer professionals.
In 1968, I was asked to edit the
Operating Systems (SIGOPS) newsletter. I got involved and quickly
toe
it up, to cover less
's need stronger leader~ should have don't
, like performance
i computer architecture."
lea
SIC
"Sp
AC
go
} is involved in ACM,
I most respected
I the computer field.
19 his career. Sharing
I making a contribution
:er profession.
tee
co
thl
on
"f\
L
c:
ing ACM. Fill out and
.
. >n today .
San Jose
Public Library
FOR OVERTIME A CHARGE OF 5 CENTS A DAY
Careful usage of books is expected and any soiling
damage or los8 is to be paid for by the borrower. '
Announcement
815 Washington St., Newtonville, Mass. 02160 invites your subscription to:
'The C&A Notebook on
COMMON SENSE. ELEMENTARY AND ADVANCED
devoted to research, development, exposition, and illustration of the most
important (or one of the most important) of all branches of knowledge, i.e.,
the subject of
WHAT IS GENERALLY TRUE AND IMPORTANT =
+
+
+
+
+
+
+
+
Purposes:
to
to
to
to
help you (and us and anybody) avoid pitfalls
to point out new solutions to old problems
to stimulate one's resourcefulness
prevent mistakes before they happen
display new paths around old obstacles
to increase one's accomplishments
apply in practical situations the observations
to improve one's capacities
and wisdom of great scientists and wise me
For additional information see in the April issue the
Editor: Edmund C. Berkeley
editorial on page 6 and the article pages 27-31.
- - - - - To:
- - - - - - - -(may be copied' on any piece of paper)- - - - - - - - - - - - -
Computers and Automation
815 Washington St., Newtonville, Mass. 02160
Yes, please enter my subscription to "The C&A Notebook on Common Sense, Elementary and Advanced"
~(
) as a READER, subscription $12 a year (24 issues)
Please enroll me~(
) as a PARTICIPATOR, subscription $6 a year (24 issues)
In order to be a participator, I AGREE to send to the C&A Notebook reasonable contributions to this project of at least 400 words per month -- of notes, comments, ideas, concepts, principles, examples, anecdotes, proverbs, etc. (Note: Only 10 days grace period!) Please send me a description of "reasonable
contributions".
Please send me as free premiums for subscribing:
1. Riqht Answers -- a Short Guide to Obtaining Them
2. The Empty Column
3. The Golden Trumpets of Yap Yap
) I enclose $_ _ __
Name
) Please bill me.
4. Strategy in Chess
5. The Barrels and the Elephant
6. The Arqument of the Beard
) Please bill my orqanization.
Returnable in 7 days for full refund if not satisfactory (if in salable condition)
Title ________________________________________________
Organization___________________________________________________________________________________________
Address (including zip) _________________________________________________________________________________
Signature___________________________________________________________ Purchase Order No. __________________
+
Vol. 20, No.5
May, 1971
COnlputers
and automation
The magazine of the -design, applications, and implications
of information process'ing systems.
Editor
Edmund C. Berkeley
Assistant Editors
Linda Ladd Lovett
Neil D. Macdonald
Software Editor
Advertising
Director
Art Directors
Contributing
Editors
Fulfillment
Manager
13
[A]
PROGRAMMING AND MINI-COMPUTER COSTS
by David R. Ellis, Director of Programming, Interdata
How to "properly" program a mini-computer; i.e. how to
implement software and documentation that yields a minimum cost solution (including performance, testing, training,
and maintenance) to some particular task.
30
[A]
PLAYING GAMES WITH A COMPUTER
by James R. Palmer, Technical Staff, Hughes Aircraft Co.
How an effective educational game to teach factory ope~ion
was designed employing a computer - and is successfully uSed
at a cost of less than $9 per team to play the game.
31
ESSENTIAL COMPUTER CONCEPTS FOR TOP MANAGEMENT[A]
by Frank J. Gabriel and the Editor
An inventory of sl ightly over 100 computer concepts, proposed In order for top management t~mputers.
19
[A]
COST SAVINGS POSSIBLE IN DATA PREPARATION
by William J. Primavera, Burroughs Corp.
A report on critical factors, including machine utilization,
error control, pre-preparation of source data, etc.
38
THE GROWTH OF THE COMPUTER INDUSTRY
by James B. Hunter, President, Digital Scientific Corp.
Stewart B. Nelson
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
Leslie Mezei
Ted Schoeters
Richard E. Sprague
Advisory
Committee
The Computer Industry
James J. Cryan
Alston S. Householder
Bernard Quint
William J. McMillan
[F]
Computers and Earthquakes
Edilor;alOffices
Berkeley Enterprises, Inc.
33
RESTORATION OF SERVICE AFTER EARTHQUAKE
by D. L. Talley, Redcor Corp.
33
UNINTERRUPTIBLE POWER SUPPLY PROTECTED NASA
COMPUTER DURING EARTHQUAKE
by Robert R. Bentley, General Electric
815 Washington St.,
Newtonville, Mass. 02160
617-332-5453
[F]
[F]
Computers and Government
AdtJertis;ng
Contact
21
[A]
MAYBE THE COMPUTERS CAN SAVE US AFTER ALL
by Edward You rdon
A proposal for a privately owned and operated National I nformation Bureau which would serve as a central source of information for anyone on any subject of reasonable interest.
37
LIFE SUPPORT AND WAR CONTROL SYSTEMS
FOR PLANET EARTH
by Howard Kurtz, War Control Planners
THE PUBLISHER
Berkeley Enterprises, Inc.
815 Washington St.,
Newtonville, Mass. 02160
617-332-5453
Computers and Automation is published monthly
(except two issues in June) at 815 Washington
.St., Newtonville, Mass. 02160, by Berkeley En·
terprises, Inc. Printed in U.S.A.
Subscription rates: United States, 11 monthly
i~sues and two issues in June (one of which
is a directory issue) - $18.50 for 1 year, $36.00
for 2 years; 12 monthly issues (without directory
issue in June) - $9.50 for 1 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: Berkeley Enterprises,
Inc., 815 Washington St., Newtonville, Mass.
02160. Second Class Postage paid at Boston, Mass.
Postmaster: Please, send all forms 3579 to Berkeley
Enterprises, Inc., 815 Washington St., Newtonville,
Mass. 02160. CD Copyright 1971, 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 allow three weeks for the change to
be made.
4
[A]
Compu ters and Socie ty
8
THE INFORMATION REVOLUTlON[A]
AND THE BI LL OF RIGHTS
by Jerome B. Wiesner, Provost, and President-Elect, Mass. Inst. of
Tech.
How an "information tyranny" could develop, unless the legislative and legal systems of this country bring up to date the
basic safeguards for protecting the privacy of individuals.
COMPUTERS and AUTOMATION for May, 1971
36
7
55
32
34
33
LYING BY THE UNITED STATES GOVERNMENT:
[A]
AN "ACCEPTABLE LEVEL"?
by Edmund C. Berkeley, Editor, Computers and Auto~a!ion
Apparently the truthful battle toll of American soldiers
destroyed by -the war in Vietnam, which the computers in the
Pentagon ought to report, is not 54,000 but over 200,000.
WHY THE PUBLIC DISLI KES COMPUTERS
by Leroy Pope, The News, Framingham, Mass.
THOMAS J. WATSON FELLOWSHIPS AWARDED TO 70
STUDENTS FROM 34 COLLEGES
by Robert O. Schulze, Thomas J. Watson Foundation
$290,000 Awarded in Libel Damages to an Insurance Broker
Suing Retail Credit Co., from Computerworld
[F]
[F]
[G]
"A Data Bank for Narcotics Addicts" - Comment
by Bruce Madsen
[F]
"To Help Liberate One's Mind from Newspeak" - Comment
by M. L. Huber and the Editor
[F]
Computers, Science, and Assassinations
27
THE ASSASSINATION OF PRESIDENT KENNEDY:
[A]
The Spatial Chart of Events in Dealey Plaza
by Robert B. Cutler, Architect
The chart, first published in May 1970, is revised and brought
up to date.
34
THE DEATH OF WALTER REUTHER
by Terrell L. Elrod and the Editor
[A]
The Golden Trumpet
54
Stop Expressing Political Polemics, by John Culleton
[G]
54
Don't Blame President Nixon
by James E. Crow and H. Irvin Smith
[G]
54
Masquerading as a Computer Journal, by Patrick M. Cooney
[G]
54
Surprise, Delight, and Support for Your Views and Concerns
by John R. Macleod
[G]
34
C&A Provokes Thin~ing, by Vincent K. Roach
[G]
55
Publishing Courage, by Robert W. Carter
[G]
55
Oasis of Truth, by Mrs. Grace P. Vale
[G]
Fron t Cover Pic ture
The front cover shows some
electronic circuit boards which
form the "backbone" of computercontrolled automated warehouses
and manufacturing plants.
Such
boards provide a link or interface
between computers and the control
systems
which
run
materials
through large factories and warehouses.
The boards translate
"slow" signals of 1/50 second from
equipment into microsecond signals suitable for computers. The
boards shown are from Cutler-Hammer Industrial Systems Division,
Milwaukee, Wisc.
Departments
Cornputers and Puzzles
Across the Editor's Desk
Appl ications
Educations News
Research Frontier
Miscellaneous
Advertising Index
Calendar of Coming
Events
Monthly Computer
Census
New Contracts
New Installations
New Products and
Services
40
40
41
42
42
53
39
50
26
Numbles, by Neil Macdonald
[C]
39
Problem Corner, by Walter Penney, COP
[C]
48
49
43
The Unemployed Compu ter Professional
6
How an Unemployed Computer Professional Might Start His Own
Business and Earn a Reasonable Income as His Own Employer [E]
by Edmund C. Berkeley, Editor, Computers and Automation
18
The C&A Notebook on Common Sense, Elementary and Advanced [N]
The Ninth Annual Computer Art Contest
fAl - Article
[Cl - Monthly Column
Announcemen ts
3
Key
[N]
[El - Editorial
[Fl - Readers' Forum
[G] - The Golden Trumpet
[Nl - Notice
COMPUTERS and AUTOMATION for May, 1971
5
EDITORIAL
How an Unemployed Computer Professional Might
Start His Own Business and Earn a Reasonable Income as His Own Employer
It seems to me that many an unemployed computer
professional could, if he wished, start his own business, and
after a period of something like two to six months earn a
reasonable income as his own employer.
He might thus attain the pleasant state of NEVER
AGAIN being able to be fired. He might even ask himself
"If I could work at whatever I wanted, what would I do?"
And he might choose work which is closely in accordance
with his deepest desires.
Important qualifications are needed however in order for
that first statement above to be true:
• Not all unemployed computer professionals, only
some, could do this (the others probably should
not try);
• The usual entrepreneurial requirements (common
sense, perseverance, keen observation, honesty,
caution, shrewdness, etc.) must be met;
• Good business practices must be learned and followed (the median life of a small business is, I
understand, about five years);
• In the early stages, both hard work and continual
close attention, usually for over 60 hours a
week, will be necessary;
• The choice of products and services to be offered
is a critical question that has to be solved well;
etc.
Contrary to what many people may think, "ample
financing" at the start of a new business - for example,
enough money to rent an office, buy furniture, and hire
several people for several months before obtaining any sales
- is I think not usually desirable. To be able to do this with
other people's money (investors' money, for example) has a
tendency to encourage wasteful and economically unsound
business operation. Bad habits for both persons and businesses are hard to change: it is better not to acquire them.
But the large number of successful small businesses
(carpenter, gift shop, salesman, copying service, electrician,
shoe repair, record and music shop, real estate office, ...)
reveals how common are the qualities of the entrepreneur,
the businessman. Even in a socialist country like the Soviet
Union, the one-man business with no employees is a legal
activity. Note that extensive formal education is often
unnecessary.
The computer profession like a computer itself has much
input and much output. Many computer professionals have
come into the computer field directly from other fields and
without much specialized education. They have previously
been in fields such as education, or management, or
mathematics, or electrical engineering, or systems design.
Similarly, a great many computer professionals may, and
will, leave the field for other occupations such as manager,
engineer, administrator, lawyer, or entrepreneur. This editorial is a plug for becoming an entrepreneur.
In fact there is little reason why many computer professionals need continue to be computer professionals. It does
not require seven years of specialized education to learn to
be a computer professional - as it does to become a
medical doctor. At least some persons with talent can
become adequate computer professionals in about two
months of assiduous learning and some practice on a
computer. And, for many people, a computer or data
processor is not hard to understand - not much harder to
understand than a machine that will knit socks, or a
machine that will weave a pattern in cloth.
If a computer person decides to change his main occupation to entrepreneur, what should he produce or offer for
sale?
There are a great many products and services whicr
consumers want, probably more than a million kinds. There
are a great many varieties of manufacturing and service
organizations which produce these products and services,
from those which consist of one man and almost no capital
equipment to those which consist of thousands of persons
and tens of millions of dollars worth of capital equipment.
However, it is not true that all the competitive advantages lie with the big organization. Many kinds of laws such
as Parkinson's Law, the Peter Principle, etc., work to the
disadvantage of the big organization. And the big organization which may be effective in year N may be almost the
opposite in year N plus 3 - after some key people have left
or retired or other changes have occurred: witness the Penn
Central.
In regard to products or services in the computer field, I
have noticed at least three items which could have a large
market. Also, so far as I know, nobody at present provides
them. These are: (1) a negative line feed (Le., a line
rollback) on typewriter-like keyboard output from a computer (such as a teletype); (2) a non-standard typing keyboard (such as the Dvorak keyboard) which could save 30%
to 50% of human time inputting information into a computer; and (3) methods of computer-assisted documentation of computer programs, which I believe could rescue
upwards of a billion dollars worth of computer programs
for which effective documentation is almost nonexistent.
Outside of the computer field, I have noticed several
dozen products and services which it seems to me that
many consumers would eagerly buy if the price were
reasonable. One example is an electric clock with a nonbreakable plastic case. This would avoid the usual built-in
obsolescence of the usual electric alarm clock, which breaks
(Continued on next page)
6
COMPUTERS and AUTOMATION for May, 1971
READERS' FORUM
WHY THE PUBLIC DISLIKES COMPUTERS
Leroy Pope
From liThe News"
Framingham-Natick, Mass.
Jan. 8, 1971
A bachelor, leaving a hospital, is handed a bill for
delivery of a baby.
A New Yorker keeps getting bills from the Motor
Vehicle Violation Bureau for a parking ticket he paid three
months ago, and each month the fine goes up $5.
An accountant loses his checkbook and the bank advises
changing his account number. Within three weeks, the bank
refuses to honor instructions to make automatic monthly
accounts from either the old account number or the new
one!
Another bank sends a statement to a depositor informing him cheerfully it has paid all ~is monthly bills and has credited his account with the total instead of
deducting it!
Literally thousands of persons across the country are
infuriated and confused each month by such mistakes
apparently made by computers. Credit card and other
billing mistakes cause the most anguish.
"Others are terribly annoyed because they constantly
see the computer as an invader of privacy that knows more
about you than you want anyone to know," says Felix
Kaufman, partner in the accounting firm of Lybrand, Ross
Bros. & Montgomery.
"But the real reason so many people hate computers is
that they are the first machines in history to really move in
on our intellectual and emotional lives," Kaufman said.
"The computer can't think and certainly can't feel. Yet
it gets involved in both intellectual and emotional decision
making - such as casting horoscopes and computerized
when it is knocked off the bedside table, and falls on the
floor. I once asked General Electric to make such a clock;
they refused.
What qualifications do I have that I might speak as an
authority on this subject? One is experience.
In 1948, 23 years ago, I resigned from a safe, salaried job
in a large life insurance company home office, where I was
doing actuarial and methods work, and I started in business
for myself, with a part-time secretary. I have now had
therefore 23 years of experience operating and running a
small business; and we have met over 500 payrolls on time.
The business has had ups and downs, and narrow' squeaks;
but I am still an eighty percent owner of Berkeley Enterprises Inc.; we have always had enough money in the bank.
to pay all our debts as they fell due. We have made a flock
of mistakes, and have learned lessons from most of them.
Certain misfortunes we have seen no way to avoid, but we
dating services," he explained.
An insurance company cancels a motorist's liability
policy on the basis of a computer "throwout." Of course
this is a human decision by an executive to throw out all of
a certain class of policyholders. "But the computer gets the
blame, and, in a sense, justly so," said Kaufman. "If the
company didn't have all its policy data stored in a computer executives would have to hunt up and weigh every
case individually before cancelling."
The computer thus becomes, in the public mind, either a
Machiavellian intelligence or a colossal blunderer. It is
neither, it is a slave that does exactly what it is told to do at
amazing speed.
Nearly all so-called computer errors are human errors. A
few are machine errors, mistakes of peripheral reading and
printing machines not the computer itself.
Take the bachelor startled at being handed a bill for
having a baby. Two human errors were involved. The key
puncher transposed the code number for his minor operation into the code number for a baby case.
"But the programmer made the original mistake. He
forgot to program an automatic red light in case an
obstetrical or gynecological charge turned up for a male
patient," said Kaufman.
People started hating computers when the banks began
using numbers instead of names to control checking accounts. Women in particular resented it, according to
Kaufman.
The bright ray in the computer cloud is the act of
programming. While relatively young and error prone, it
should get better. "Better programming," says Kaufman
"eventually will reduce errors and overcome public dislike
of the computer."
have adapted to the limitations they have placed upon us.
If any computer person who reads this editorial is a
subscriber to "Computers and Automation", is unemployed, is seriously interested in trying to develop a small
business of his own, and is willing to send us certain data
about his aptitudes, we will send him:
(1) a list. of several dozen products and services for
which we have observed a need;
(2) some further remarks and guidelines drawn from
our experience and thoughts on the subject of
operating one's own business.
For further information, see the details of this offer on
page 29.
Edmund C. Berkeley
EDITOR
COMPUTERS and AUTOMATION for May, 1971
7
THE INFORMATION REVOLUTION
- AND THE BILL OF RIGHTS
"The great danger is that we could become 'information bound', because
each step in the development of an 'information tyranny' appeared to be
constructive and useful. I suspect that it would be much easier to guard
against a malicious oppressor than to avoid being slowly but increasingly
dominated by an information Frankenstein of our own creation. "
Dr. Jerome B. Wiesner, Provost and President-Elect
Massachusetts Institute of Technology
Cambridge, Mass. 02139
The way in which we use and control the great new
capabilities being created by the information revolution will
shape the future character of our society; it may be said,
indeed, to be shaping it already. Technology has been
providing mankind with new tools and new opportunities
for a long Line and, in response, society has evolved new
institutions and has changed its physical form. Sometimes
these responses have been comfortable and swift, as in the
case of the telephone and radio, at other times, they have
been halting and painful, requiring repeated trials with
many errors to find a new eqUilibrium that was comfortable
for the socie ty.
Knowledge IS Power
For a long time, the rate of technological progress was
sufficiently slow to enable society to adapt to the required
change without permanent distortion of values. The pace of
change is now very swift. We say "time is shorter now", and
that is why we are faced with our present problem. To
make the matter particularly urgent, information threatens
to undo that subtle balance achieved in the Constitution
between the people and the state which avoids anarchy on
the one hand and tyranny on the other. Nowhere is it more
true that "knowledge is power". Information technology
puts vastly more power into the hands of government and
the private interests that have the resources to use it. To the
degree that the Constitution meant for the power to be in
the hands of the "governed", the widespread collection of
personal information poses a threat to the Constitution
itself. There is also no doubt that technology can be and
has been used to assist in the violation of the Bill of Rights.
But it must be remembered that the violations are made by
humans, not by machines. To my non-legal mind, there is
even the question of whether the Bill of Rights, drafted in a
Simpler time, is adequate to protect man in his relation to
the modern state and, whether there isn't a need for
additional amendments providing protection for the individual against possible new infringements of his liberties.
Based on a statement delivered to the Constitutional Rights SubCommittee of the Senate Judiciary Committee on March 11, 1971.
8
Because many of our difficulties stem from the unforeseen side effects of technology or from the misuses of
technological capabilities, there is a growing resentment and
antagonism toward science and technology. There is also a
widespread feeling that mankind would be better served if
we could retreat to a simpler time. Given the present size of
the world's population and the complexities of modern
society, this hardly seems possible. In fact, I am firmly
convinced that only through the sophisticated and careful
use of technology can we create a truly decent society. In
this circumstance, we must learn to manage technological
change effectively for the common good. This, it seems to
me, is the particularly important and urgent task of the
Congress. Many committees of the Congress are concerned
with aspects of this problem (such as the present hearings
on the SST), but there is little focus on the overall task.
1984 Could Come Unnoticed
Modern information technology provides the potential
to add to our general well-being and to enhance human
freedom and dignity if properly used by extending our
muscles, brainpower and material resources. Yet it also
threatens to ensnarl us in a social system in which controls
could essentially eliminate human freedom and individual
privacy. Improperly exploited computer and communication technology could so markedly restrict the range of
individual rights and initiatives that are the hallmark of a
free society and the foundations of human dignity as to
eliminate meaningful life as we appreciate it. In other
words, 1984 could come to pass unnoticed while we
applauded our technical achievements.
The great danger which must be recognized and counteracted is that such a dc-personalizing state of affairs could
occur without specific overt decisions, without high-level
encouragement or support and totally independent of
malicious intent. The great danger is that we could become
"information bound", because each step in the development of an "information tyranny" appeared to be constructive and useful. I suspect that it would be much easier to
guard against a malicious oppressor than to avoid being
slowly but increasingly dominated by an information
COMPUTERS and AUTOMATION for May, 1971
Frankenstein of our own creation. (Though we should
recognize, I believe, that an effective means of citizen
surveillance and intimidation could also provide attractive
opportunities for a would-be dictator.)
Control of Information
Present and growing capabilities for surveillance and
control are made possible by modern communication and
computational techniques. It is very clear that such capabilities, through data-centralization and manipulation, will
continue to grow at an ever increasing rate as our understandings of communications, computation and cognitive
processes expand. At the same time, it is obvious that
means for effective record keeping, information gathering,
and data processing are essential needs of a modern society.
The problem is to determine how to reap the maximum
assistance from modern technology in running a better
society and at the same time, how to keep it from
dominating us.
In order to do this, we may have to adopt some stern
measures in the form of very strict controls on who can do
what with private information about any individual in the
society. The present capabilities in information collection
have already lead to clear-cut infringements of citizens·
rights. In fact, even without technological assistance, there
have been serious violations of the Constitutional protections by many agencies of the government and by many
private organizations. Furthermore, the awareness of security dossiers has inhibited many people in their political
activities.
What Kind of Society Are We Building?
There is one specific point which I would like to stress.
The issue of constitutional rights is but one dramatic aspect
of the major problem of our time; namely: given so many
options by a rapidly developing technological capability,
what kind of a society are we going to allow to be created
for ourselves and for our descendents? We live at a moment
in history - I believe a unique moment - when the
decisions we make, the paths we take, will shape the future
of man's world for a long time to come. Technology allows
us exciting opportunities for shaping a world to our liking,
but it also poses the possibilities of a disastrous misstep.
People everywhere have begun to appreciate that the
thoughtless applications of technology on a large scale,
done with the best of intentions and for the most constructive purposes, can frequently have large-scale destructive at least, very unpleasant - side effects. We have slowly
come to realize that we can intervene into the workings of
the physical world on a scale and in ways that actually
threaten man's survival on this planet: Fortunately, there is
a widespread reaction against such careless actions;witness
the growing concern for the environment and the growing
disenchantment with war, particularly nuclear war, as an
instrument of foreign policy.
The Effects of Small-scale vs. Large-scale Applications
I
We are also beginning to understand that we can affect
man's social and psychological environments in equally
disturbing ways. We have learned one particularly important lesson about all of this. It is that technological
innovations that are wholly constructive when employed on
a small or moderate scale can, with increased and constant
COMPUTERS and AUTOMATION for May, 1971
application, have such serious impact on the environment
or on the society that massive efforts are required to offset
their disastrous side effects. Sometimes a technical innovation can affect both the physical and psychological environments. The automobile, electric power and the aeroplane all
illustrate this point.
The early manufacturers of automobiles hardly anticipated that their machines would produce the Los Angeles
smog, the blight of our cities, or the malaise of the suburbs.
And even today, the individual user of a bit of technology
such as the automobile, a pesticide or a polluting detergent
clearly believes that his personal gain greatly outweighs the
environmental hazards that his small transgression produces. On the other hand, it is perfectly obvious that
citizens of our country are sufficiently concerned about
these problems to be willing to legislate against pollution
even at the price of considerable inconvenience and cost.
They are ready to spend substantial sums of money for less
destructive products and large sums to undo the environmental damage from the past.
Weapons Control
An important lesson can be learned too from our efforts
to control weapons systems. It is much easier to stop the
application of a specific piece of technology or a specific
technique or a new strategy before it has been developed or
widely applied, than after the fact. For example, it was
relatively straightforward for the United States and the
Soviet Union to agree to prohibit the introduction of
nuclear weapons in outer space or on the ocean sea beds
because these weapons did not exist. In contrast, it has
proven impossible to curtail the emergence of anti-ballistic
missile systems, even in the face of widespread agreement
that they can't be effective, because they do exist and the
decision to halt their development and deployment is
contrary to the interests of large groups of people. It is
perfectly obvious that this is a generally applicable theorem. If we want to avoid traumas from the 'mis-application
of technology in the future, we should learn to recognize
the inherent environmental or social threats in an early
stage of a new technological development. I think that this
point is particularly important in the matter of preserving
privacy and freedom. The motto "eternal vigilance is the
price of liberty" applies here with special meaning.
Surveillance of People
Modern electronic aids are not required for the operation of a comprehensive surveillance operation. In fact, the
very effective security systems run by the Defense Department and the FBI during and after World War II made only
modest use of electronic information storage and retrieval.
But such systems were consequently quite expensive and
also limited in the number of people that they could watch
over closely. They freEiuently bogged down when presented
with too much information. Large-scale data systems now
operated by government bureaus and even private credit
bureaus maintain files on tens of millions of people with no
difficulty whatsoever. Furthermore, as you know, interconnecting communication networks allow information in separate files to be coordinated and centralized with great ease.
In addition, as the software for data analysis becomes more
sophisticated, it will be possible to simulate patterns of
behavior for individuals and social groups and attempt to
predict or anticipate their behavior with the purpose of
9
maintaining better surveillance on individuals who, in one
sense or another, might represent a threat to someone
having access to the data system.
In a way, this was what the Army was doing when it
selected anti-war organizers, speakers and demonstrators for
particular attention. Apparently, their system was not very
effective and it was detected before it could become a
substantial threat to the freedom of the individuals involved
and to the nation as a whole, but nonetheless, it is an
indication of the threat to privacy and freedom that does
exist.
Who Is Under Surveillance?
In fact, I wonder if the full extent of that Army
surveillance system has really been exposed? When I was
President Kennedy's Science Adviser, I found that it was
frequently very difficult to find out the full scope of similar
organizations. I have participated in anti-Vietnam war
demonstrations, I have spoken at them, I have marched in
them, and I have wondered lately whether I am being
watched as a threat, as a dangerous enemy of the realm.
How do I know if I am under continuous surveillance?
The answer is, I don't know. I doubt that anyone is
aware of the full extent of the surveillance and information
collection activities that go on in this nation.
Many people, myself included, have long operated on
the assumption that our activities are being monitored. I
have also operated under the premise that I should not
allow myself to be inhibited by such a possibili ty. I do this
because I have great confidence in the basic integrity of the
safeguards built into the administrative and judicial system
of the country. If I lacked such confidence and did not feel
that I could defend myself, were there to be unjust
conclusions or accusations, I would undoubtedly feel much
more severely restricted.
Students Do Feel Inhibited
I know that many, many students are afraid to participate in political activities of various kinds which might
attract them because of their concern about the consequences of having a record of such activi ties appear in a
central file. They fear that at some future date, it might
possibly cost them a job, or at least make their clearance
for a job difficult to obtain. I don't kno~ to what extent
these student fears have any justification, but I can tell
you that they are real fears and that they frequently have
caused students to back away from activities which attracted them. I might add here that I am not referring to
confrontations or planned violence, but participation in
seminars, political study groups, etc., that were seriously
questioning governmental and social arrangements or policies.
Technical Safeguards Are Not Enough
There are those who hope that new technology can
redress these invasions of personal autonomy, existing or
prospective, that information technology makes possible,
but I don't share this hope. To be sure, it is possible and
desirable to provide technical safeguards against unauthorized access to data banks or information transmission
systems.
It is even conceivable that computers could be programmed to have their memories fade with time and to
eliminate specific identity when the information was being
10
processed to provide social profiles, etc., and such safeguards are highly desirable, but the basic safeguards cannot
be provided by new inventions. They must be provided by
the legislative and legal system of this country. We must
face the need to provide adequate guarantees to individual
privacy.
Specific Needs
I am a communications specialist, not a legal expert, and
consequently, I hesitate to propose specific legislation.
However, I have spent considerable time thinking about the
issues involved and I would like to mention several specific
needs which I see. These are:
1) A watchdog authority, perhaps an independent
agency, possibly a division of the General Accounting Office, perhaps the FCC, to review regularly the public and private information gathering
and processing activities within the country. The
agency should have the authority to examine the
nature and extent of such activities and should
report its findings to the Congress and the public.
2) Congress should set rigid limitations on permissible
surveillance activities and establish much stronger
safeguards than now exist against misuse of datafile information.
3) Action should be taken as quickly as is feasible to
re-establish public confidence in the sanctity of
the boundaries of an individual's physical and
psychological living space. This will require a
number of steps. Outlawing some activities such as
the free exchange of private information, collecting data not needed by an agency, etc., will help a
good deal. Acknowledging publicly the extent of
permissible surveillance and by whom is also
important. Requiring disclosure of non-security
type data to the concerned individual seems possible in many situations. In the few situations where
this will not work, as in national security matters,
judicial controls should be strong.
4) Technical means of insuring data security and
safeguarding privacy should be developed vigorously and their use required.
A Balance Between Threats to Freedom - and Freedom
We should be prepared to accept the cost of considerable
inefficiency in our various social and governmental pro-.
cesses to safeguard our privacy and, as I judge it, our
freedom, dignity, happiness and self-respect. By costs, I
mean both the financial costs and the loss of a degree of
control that the state might otherwise have over genuinely
threatening individuals such as criminals and violent revolutionaries. Our task is to achieve a proper balance between
the ability to cope with individual threats to the society
and its capability to abridge the freedom and happiness of
its members. In countries where the legal system cannot be
counted on, the people are at the mercy of the administrators and they must hope that the bureaucracy will be
benign. Such a situation smothers freedom. Because I
believe that an "information tyranny" poses a very serious
threat to the survival of a free society in our country, I
vigorously recommend that Congress take whatever steps
are necessary to bring the Bill of Rights up to date.
0
COMPUTERS and AUTOMATION for May, 1971
PROGRAMMING AND MINI-COMPUTER COSTS
"Computer ads and brochures often supply many words and few useful facts
about software. Documentation on the subject often uses jargon and
symbolism that makes it dlfficult for the novice to read. In fact there are few
criteria available today to enable anyone to objectively compare the quality
of existing software. Yet the dlfference between a costly and a cost-effective
use of mini-computers depends on software. "
David R. Ellis
Interdata, Inc.
2 Crescen t Place
Oceanport, N.J. 07757
Minicomputers are gaining in popularity and acceptance
at a phenomenal rate. One reason is the apparent getsomething-for-nothing appeal that stems from computers
with greater and greater power getting less and less expensive. While it is true that computer hardware costs are
falling at a rapid rate because of technological advances, it
is not always true that the cost of putting these computers
to use is falling proportionately. In fact, the opposite is
often the case.
Minicomputers of today provide the engineer an unprecedented opportunity to implement or improve an
endless variety of data collection, automatic test, and
control tasks which otherwise would be handled in a more
expensive, manual fashion which is subject to errors. The
difference between a costly and a cost-effective use of
minicomputers for such applications depends on programming. Minicomputers must be programmed to handle each
specified application with specific software. However, the
hardware characteristics of many minicomputer systems the computer architecture, memory size, and peripheral
devices - do not lend themselves to easy generation of
software. In fact, some minis are downright difficult to
program. When applying a mini therefore, the application
engineer must give careful attention to programming requirements.
David R. Ellis received a BSEE degree from Purdue
University, and an MSEE and EE degree from the
Massachusetts Institute of Technology. He joined
Interdata in 1967, and is presently the Director of
Programming. He has been active in the development
of operating systems and systems software for minicomputers since 1964.
COMPUTERS and AUTOMATION for May, 1971
This article discusses the minicomputer software situation and sets down some guidelines for proper progamming.
The phrase "proper programming" means implementing
software and documentation that yields a minimum cost
solution to some task, the specifications for which must
cover operating performance, testing, training and maintenance requirements.
Sources of Software
The prospective buyer or user with an application
problem to solve must ultimately face up to the question:
How do I obtain the software necessary for my application?
From a computer vendor? From an independent supplier?
Or must I write the programs myself? Before trying to
answer the questions, let one fact be made clear - the low
dollar figures for minicomputers that are quoted in advertisements and news releases do not normally include the
specific application software needed to solve your particular problem.
System Software
What is normally included in the price of the hardware is
a set of the appropriate "system software" programs associated with that computer, such as:
a. Symbolic Assembler
b. Fortran Compiler (if any)
c. Other language processors (like Algol, Cobol, if
any)
d. Operating Systems (i.e. DOS, RTOS, if any)
e. Math Library
f. Utility routines (i.e. program loader, debug program, editor, conversion routines.)
Another type of system program is a diagnostic or test
program. Such programs are not normally associated with
any application, nor are they used in the software development s'equence.
13
The system software programs do not handle any application problems by themselves. Rather, they serve as tools
which help the user to create the types of programs which
can be used for specific applications. For example, the use
of a minicomputer for process control requires additional
programs which read the analog or digital inputs at a proper
rate, perform control calculations appropriate to the process being controlled, and generate proper control signals to
the valves or actuators which control the process. These
programs could be generated with the assistance of the
system software programs by someone who understands the
process control environment.
A program preparation sequence using the assembler
might run as follows:
a.
b.
c.
d.
e.
f.
g.
h.
i.
Enter bootstrap loader into memory.
Enter program loader into memory.
Load the editor program.
Use the editor program to create source tapes of
the user's programs.
Load the symbolic assembler.
Assemble the source tapes, which convert the
symbolic language into binary machine language,
punching binary object tapes.
Load the assembled binary object tapes into
memory.
Test and debug the object programs using a debug
program.
Repeat steps as necessary until the new programs
are satisfactory to the user.
An exception to the statement that system programs are
always provided with the minicomputer hardware occurs
with those manufacturers that have unbundled, which
means simply that they charge for the system software
separately from and in addition to the charge for the
hardware. Very few mini manufacturers are unbundled, but
the buyer must watch for separate pricing policies.
2.
3.
4.
5.
6.
An obtainable application package might provide a complete program which performs some entire end-user function, or it might provide a partial solution to a problem,
requiring the buyer to make extensive changes or additions.
It is seldom clear from the designation or the descriptive
literature just how complete an application package really
is, or how it is tailored to one's needs. Application software
claims need to be scrutinized carefully. To procure the
proper application software, you have to know what obtainable packages are available, as well as the considerations
involved in developing the system yourself.
The Development Process
Here are the steps to follow in a typical development
process for a minicomputer application system:
1.
2.
3.
4.
5.
6_
7.
S.
9.
Application Programs
To adapt a mini to your specific application, then, you
must acquire application programs. Whether these should
be purchased from the vendor, purchased from an independent organization such as a software house, or written
by yourself depends on many factors. When dealing with
the manufacturer, one factor is how many computers you
will need. If several machines are involved, the manufacturer may be interested in producing your application
software. On the other hand, if only one computer installation is involved, the manufacturer may not consider it
economically feasible to write specific application software.
Another significant factor is how closely your application
conforms to what other people have done in previous
situations. If your job, or a very similar one, has been
solved many times before, you stand a much better chance
of finding an interested supplier who will provide the
needed software at a reasonable price.
Pre-Canned Packages
There are also companies who offer pre-canned packages
that claim to solve certain general problems. Examples of
application packages are as follows:
1.
14
IDACS-S from Digital Equipment Corporation
for industrial data acquisition and control.
DC 2 from Fisher Controls Corporation for process control applications.
Key Logic from Redcor for keypunch to disc
data collection tasks.
FFT from INTERDAT A for on-line Fast
Fourier Transform calculations.
HP 2000 Series from Hewlett Packard for general purpose in-house time sharing systems.
The 270X from INTERDATA for 360 frontend communications handling.
10.
Define the system and the project.
Evaluate vendors.
Acquire minicomputer hardware.
Train or acquire the development staff, if necessary.
Acquire a special development system, if necessary.
Design, write, assemble and test the software.
Document the software (for training, operation
and maintenance).
Install and test the system.
Train users of the system (on operation and
maintenance ).
Write project reports for follow-up on other
development projects as needed.
In reality, the steps in such a process are iterative in
nature, and overlapping in time. For discussion purposes, it
is helpful to show them as discrete steps which proceed
sequentially. If either the entire application system or the
application software is purch~sed, rather than just the
minicomputer hardware, then the development process
changes somewhat. (See Figure 1.) The actual purchase of
the minicomputer hardware (step 3) is but one step in ten.
The cost of the hardware may in fact be a small part of the
total cost, depending on how much time and effort is
required to satisfy the other nine steps. While minicomputer prices have tended to shrink, the cost of people's
time or talent has increased; as a result, the ratio of total
system costs to hardware costs has risen dramatically.
To illustrate, consider a hypothetical application of a
modest-configuration mini for automatic testin~ purposes
in an industrial plant. The minicomputer system, assuming
no large peripheral or bulk-storage devices are involved,
may cost approximately $25,000. To estimate the total
system cost, it is necessary to estimate the total amount of
labor required for the development process. The system
definition, vendor evaluation, and hardware purchase may
be accomplished with a few people - let's say approxiCOMPUTERS and AUTOMATION for May, 1971
Figure 1 - Application Development Process
mately six mari months. The software costs, for training the
programmers, implementing and documenting, may be large
or small depending on the number and experience level of
the people involved. Let's assume that two programmers are
involved who need one month each of training, two months
each for implementation, and three months each for documentation. Total software costs would then be approximately 12 man months.
Installation, Testing, and Training
Purchase com lete
application system
Acquire
mini-computer
hardware
Purchase applications softwar
ITrain or acquire
dev~ ~~~~~~-=-~taff
No
I
Acquire s p e c i J
I l
I
development
--Yes-system as needed
I
~N~o
Then comes the installation, testing, and training of
users of the system. In an industrial operation, these tasks
may involve many people, particularly if the automatic
testing relates directly to company operations. For example, 10-15 people may be involved for two to three months,
which means that from 20-45 man months might be
required. Let us assume for this example that 32 man
months is a good estimate. Total labor required, then, is
6 + 12 + 32 = 50 man months. These figures are summarized in Figure 2. Assuming for the sake of round numbers
that labor costs $1000 per month, then the development
costs for the system are approximately $50,000, or twice
the cost of the hardware.
What does all this have to do, you might wonder, with
"proper programming." In the above example, the estimated software development costs are only $12,000. However, the labor expenses for the installation, testing and
training of the system, which amounts to $32,000 are a
direct function of the quality of the programs and documents produced during the software development phase.
Therefore, proper programming in the first stage is essential
to minimize the TOTAL system development cost, not just
the software development cost.
It is worth noting that the software development effort
follows (or should follow) the definition phase of a project.
A good application and project specification should define
the requirements for each phase of the project. For example, a need for good training material and tutorial documents for use by manufacturing personnel should be
identified at the outset of the project, since this defines the
type of software documentation required. A good definition, therefore, is fundamental to the success of a project.
It is certainly necessary before a computer can be programmed properly for a particular application.
Programming Considerations
Train users of
the system
Operate and
maintain
system
COMPUTERS and AUTOMATION for May, 1971
Computer ads and brochures often supply many words
and few useful facts about software. Documentation on the
subject often uses jargon and symbolism that makes it
difficult for the novice to read. There are few criteria
available today to enable anyone to objectively compare
the quality of existing software. Here are some guidelines
for preparing programs properly.
Software Purchase
The first point, at the risk of repetition, stresses that all
good programs, even those that are purchased, must start
with a good project definition. Even when software or an
application system is purchased, it must be installed, tested
and maintained once it is delivered. Therefore, the specs
supplied to the vendor must define what level and quality
of documentation is required to satisfy the needs of the
users after delivery. Failure to specify this information at
the time a purchase is negotiated can lead to later complications for both the buyer and the vendor.
15
Figure 2 - Estimated labor for development
of hypothetical industrial test system
Estimated No.
of People
Estimated
Man Months
Required
Defin~ system and}
2
6
Train programmers}
Implement software
Document software
2
2
4
6
proJ ect
Evaluate vendors
Buy hardware
Install system
Test system
Train users
}
10-15
32
50
largest amount of programs and data that must co-exist in
memory at one time. Note that some minicomputers use a
portion of core memory for dedicated functions, such as
interrupt pointers. The area in memory reserved for these
purposes is not available for program storage, and should be
deducted when estimating the amount of memory available
for programs and data.
If the right memory size for the end use is insufficient to
allow program preparation, then some other method must
be found to develop software for the application system.
Some alternatives in this case are:
1. Buy an additional system for development purposes only.
2. Buy enough time on another computer of that
model to develop the programs.
3. Use an assembler, if one is available, that runs on a
360 or some larger machine for purposes of development. (This approach may prove satisfactory for
program assemblies, but program debugging may
be difficult or impossible.)
4. Buy the necessary software outright.
Qualities of System Software
Hardware Selection
The prospective purchaser of a mini is confronted with a
bewildering array of vendors, and a wide range of products
from which to choose. Often the cost tradeoffs for various
items in the minicomputer product line are not obvious.
For example, should the system have 8 KBI or 16 KB of
memory? Should the buyer consider magnetic tape rather
than paper tape? Is the Multiply/Divide option worth the
money? There are no universal answers to these questions.
It IS generally true that the smaller the hardware configuration, the more difficult to develop programs, and the more
dollars required to write the necessary software. For example, if the application program requires about 8 KB of core
memory, the buyer might opt for this memory size in an
effort to save $3K to $4K in the cost of the computer,
rather than buy a 16 KB memory. However, the 8 KB
memory may not support the compiler for that computer
model. Therefore, all programs will have to be written in
assembly language, which may require additional training
for the programmers.
Furthe'r, to get the program into 8 KB may require
extensive optimization and recoding of the programs. If
more than three to four man months of programming effort
results from the choice of this memory size, it costs the
buyer more money in the long run. The same applies to
selection of a Multiply/Divide option, or Floating-point
option, or other instruction set options that may be
available. In general, scrimping on the cost of the hardware
can be penny-wise and pound-foolish for the buyer, unless
many machines are required.
Determining the best amount of memory for a particular
application can be a difficult problem. One criterion is to
get enough core to support the system programs (such as
the Fortran compiler) on the chosen minicomputer.
Another criterion is to get enough memory to support the
1
KB means kilo byte, where byte is an 8-bit unit of memory
16
Most mlm vendors supply a fairly standard set of
systems software programs with the computer: assemblers,
compilers, program loaders. Once the software development
begins in the chosen mini, these programs become very
important, since they are the tools with which the programmers must work to implement the desired result. While
difficult to measure quantitatively, there are good and bad
characteristics of system programs, and they should be
assessed during vendor evaluation. The software characteristics to examine are as follows:
1. Documentation. This is the means by which your
programmers will" learn how to use the system. The
documents should be clear and concise. For the
novice programmer, some tutorial material should
be available. The operation of system programs,
and the computer for that matter, should be
straightforward in design and easy to use.
2. Performance. The- programs should be well-tested
and perform consistent with their claimed operating characteristics. Evidence of testing may be
hard to come by, but if the performance of a
compiler is of interest to a user, it is not unreasonable to ask the vendor how it was tested.
3. Adaptability. Every user encounters needs to make
some minor modifications to some programs. The
changes may concern the use of some unique
peripheral device, or the addition of some feature
of particular interest. For this reason, programs
provided by the manufacturer should be readily
modifiable by the users themselves. To facilitate
changes, source tapes or cards and listings should
be available for all programs. In addition, program
documents, such as listings, should be written
clearly so that the program structure is explained.
Device independence, which results from the use
of an I/O monitor or operating system, when one
is available with the computer, may obviate the
need for changing many programs. Examples of
such operating systems are DOS on the Data
General Nova and Supernova, RTOS on the
COMPUTERS and AUTOMATION for May, 1971
INTERDATA Model 5, and the 2005A Real-Time
Executive for the Hewlett Packard 2116.
4. Vendor Support. Most programs provided by computer manufacturers are subject to revisions, changes
or improvements as more and more users have an
opportunity to use the program. It is the responsibility of the manufacturer to inform the customers of
these revisions and changes. Some companies convey
news regarding changes or improvements through a
newsletter to their user group.
Programmer Training
If programmers must be trained to use the new mini
then the effort required to explain the new machine should
be explored. This is another aspect of computer evaluation
which is difficult to quantify. An important factor is the
availability of some well-written training material slanted
toward the novice programmer. More important than this,
however, is choosing a minicomputer with a straightforward
design. Intricate complexities in a machine, such as fixed or
relative memory paging or an involved addressing scheme,
can slow down both learning and software development
efforts.
Another factor in training concerns the type of documentation required for the project. Traditionally, English
has been a much more difficult language to master than
Fortran, and minicomputers offer no simple solution to the
documentation problem. If well-written documents are
required to explain the application system to its users, then
it is important to get someone on the project who can
write, as well as program. This move may require some
extra training effort at the start of a project, but the
long-range benefits to the users make this wise.
Choice of Language
Assembly language is still the cornerstone of most
minicomputer software systems. For many functions, such
as interrupt handling, input/output procedures for unusual
devices, or optimatization for time-critical cases, assembly
language is essential. In almost every application, some
assembly language programs are used, and knowledge of the
assembly language is required.
Some higher-level, or problem-oriented languages, such
as Fortran, Basic, or Algol, are available, but not many
others are in use today on minicomputers. The lack of
numerous problem-oriented languages is a major distinction
between minicomputers and larger computers such as IBM
360, CDC 6600, B5500.
The need for Fortran or an equivalent language on the
mini is critical when the application program involves many
arithmetic calculations. The job of programming an arithmetic expression such as:
the mini's system software allows the user to intermix
programs written on different language levels. Also., it pays
the buyer to get those machine features, such as floatingpoint arithmetic instructions, that assist both the programmer and the system software itself in getting the job
done efficiently. Floating-point instructions, for example,
make the compiler more efficient, the arithmetic programs
much smaller, and execution times much faster.
Optimization
In programming, memory space and execution time
represent tradeoffs in a program design. That is, the smaller
a program is made, using subroutine or interpretive techniques, the slower it will execute. Similarly, for faster
execution times, more memory space is needed. Without
delving into the theoretical justification for this ph~nom
enon, let the user be aware that techniques which attempt
to nest or pack programs too tightly into core memory tend
to slow down program execution. A caution flag should be
waved if a programmer advocates a special technique such
as interpretive coding for packing more and more programs
into less and less core. Another disadvantage of excessive
progra~ packing (space optimization) is that program logic
become.s difficult to follow, difficult to test, difficult to
document, and difficult to change. In general, the best
approach is to insure that sufficient core is available to
write straightforward programs that do not employ tricks
or devious methods.
For time optimization, meaning minimizing the execution time, it is best to write programs in assembly language
using straight-line code, with as few branches and subrou tine calls as possible.
Testing the System
Thorough testing of a system is often necessary, but not
always possible. A communications concentrator system,
for .example, may not get all facets of the program testing
untIl the system is functioning on-line. At that time, it may
not be possible to generate at will all possible error
conditions, so the program's error detection and recovery
may go untested. One alternative to this situation is to
build special test equipment which simulates the real-world
?rocess. T~e testing procedures should be defined as early
In the project as possible. One technique is to build trace
points, loop counters, and special debugging messages into
the program to aid in the testing process. These can be
removed later to make the debugged program smaller.
Where testing is difficult, dealing with a vendor that knows
the application area, and has working installations in the
field, can be a big help.
Program Maintenance
Maintenance activities include:
a.
b.
using 16-bit fixed-point add/subtract/multiply/divide instructions can be an enormous job, particularly if much
numerical precision is required. For problems involving
such calculations, use of a higher-level language is a must.
Because many application systems involve both arithmetic calculations and I/O peculiarities, it is important that
COMPUTERS and AUTOMATION for May, 1971
c.
The analysis of the system for service purposes
if the system should develop a malfunction.
The correction of the system if an error in its
design is uncovered.
The changing of the system if a new or different
operating feature is desired.
Such maintenance functions all require a knowledge of how
the program was designed and implemented. Maintenance
functions also require knowing the mechanics of making
17
minor changes in the system. Depending on who will
maintain the system, and under what conditions, maintenance documentation may be trivial or it may be monumental.
In general, a clean, well-commented listing which explains the internal design of the program is essential to
perform any maintenance. Other documents, such as flow
charts, diagrams and descriptive information may be required, depending on the application.
A final thought on maintenance is that the use of a
minicomputer with straightforward architecture and simple
addressing mechanisms does more to simplify the maintenance function than all the flow charts in the world.
Software: Make or Buy
Again, proper programming requires at the outset, a
good project and design spec that defines the requirements
for system performance, testing, training, and maintenance.
Given the existence of such a specification, the prospective
user of a minicomputer system has to determine what
vendors can satisfy his needs - either with a complete
"turnkey" system, or an application software package to
supplement some standard mini, or just a plain mini with its
library of system software to help the user write his own
application programs. Even if some vendors offer a complete system, or an application software package that meets
the requirements, the user should consider the cost trade. offs of making or buying the software. Figure 3 provides a
summary of factors which affect the make-or-buy decision.
Figure 3 - Factors for Make or Buy Decision
If all the conditions in the right-hand column of Figure 3
are satisfied, the user should buy the system from a
qualified vendor. If all the conditions in the left-hand
column are applicable, then the user should develop the
software himself. If some conditions on the right and some
on the left are met, then the prolipective buyer must
evaluate the factors and establish priorities before making
the decision.
Conclusion
Software needs are of prime consideration when evaluating a minicomputer application. Falling prices of minicomputer hardware do not necessarily mean that the
software costs are falling; in fact, the converse is often true.
Programming is getting easier, due to more and better
system software available from vendors, but there are
potential pitfalls in selecting a vendor. Application programs with fancy names, or sophisticated operating systems
may not help you - the end user - at all. Therefore, to
minimize software costs, choose a mini that has good
programming characteristics, has good quality system software available, and an established company behind the
product, for support. Making that choice can get the
programming off to a good start. The proper programming
effort then requires the user to fully specify all the phases
of the job to be done - from developing, to training, to
operating, to maintaining the system. Good programs are
those which satisfy all needs, both short term and long, at
minimum cost.
D
You are invited to enter our
~~d
loose application specs
precise applications
specs
specs may change with
time
specs not likely to
change
available application
packages not similar or
need extensive modification
appropriate system
available from outside
organization
programming talent in
house
no programming talent
in house
a. hardware good for programming development
b. hardware not good for
development
i. 360 support avail
able
ii. development system available.
hardware configuratior.
not good for programming development
complete testing not
easy
complete testing is
practical
widespread or long
term use in company
anticipated for training, operation, and
maintenance
little use or interaction expected with
organization
other future application jobs likely
no jobs like it now or
in near future
18
COMPUTER ART CONTEST
the special feature of the
August, 1971 issue of
computers
and automation
GUIDELINES FOR ENTRY
1. Any interesting and artistic drawing, design or sketch made by
computer (analog or digital) may be entered.
2. Entries should be submitted on white paper in black ink for
best reproduction. Color entries are acceptable, but they may
be published in black and white.
3. The preferred size of entry is 8Y2 x II inches (or smaller); the
maximum acceptable size is 12Y2 x 17 inches.
4. Each entry should be accompanied by an explanation in three
to five sentences of how the drawing was programmed for a
computer, the type of computer used, and how the art was
produced by the computer.
There are no formal entry blanks; any letter submitting and describing the entry is acceptable. We cannot undertake to return
artwork, and we ask that you NOT send originals.
The winning entry will appear on the cover of our August issue
- more than 25 entries will be published inside, and other entries
will be published later in other issues.
Deadline for receipt of entries in our office is July 2,1971.
COMPUTERS and AUTOMATION for May, 1971
COST SAVINGS POSSIBLE IN DATA PREPARATION
"Some critical factors affecting the costs of data preparation are: operator
productivity; machine utilization; error control; scheduling; pre-preparation
of source data; and verzfication by computer using logic checks. "
William J. Primavera
Bostelman Associates, Inc.
654 Madison Ave.
New York, N. Y. 10021
Computer users of all sizes may save 20 percent to 30
percent or more of their current keypunching costs by
increasing productivity, reducing verification, improving
scheduling, or acquiring new equipment.
Although computers, systems and programming have
been continually refined to their current level of sophistication, data preparation has largely been neglected. Many
companies still prepare and enter input as they did in the
early days.
I nefficient Practices
Several typical inefficient practices include the following:
•
Keypunch operators leave their machines to get
small, unscheduled batches of work.
•
Supervisors are unaware of individual performance (since it had never been measured) and
of "normal" production (since there have been
no established standards).
•
Some formats force an operator to search for
data which is out of sequence.
•
The company applies a policy of 100% verification despite the noncritical nature of most
information.
Rising Costs for Data Preparation
While each succeeding computer generation has decreased cost per computation about tenfold, data preparation costs have actually risen. The main reason is the extra
personnel expense incurred to process the burgeoning input
that today's faster and more productive computers can
handle. Management is often unaware of the opportunity
for cost reduction and control in data entry. Take, for
example, the following three broad observations recently
made:
1.
2.
3.
Data entry costs (keypunch and closely related
activities) can represent 20 percent to 50 percent of total recurring expenses for electronic
data processing. These tend to rise faster than
processing costs.
Two critical factors are operator productivity
and machine utilization, which affect the quantity per dollar of the EDP "product". Two
more critical factors are error control and
scheduling. These affect the accuracy and timeliness of the infbrmation, the quality of the
"product".
Equipment alone does not solve data entry
problems. Potential "trade-offs" with source
data preparation preceding data entry and computer processing following data entry must be
investigated. Formats, procedures, and managerial controls must also be scrutinized. In
fact, careful analysis of the critical factors
affecting data entry may help achieve significant economies without any equipment changes
at all.
(Based on a report made for the "Lybrand Newsletter", March,
1971, published by Lybrand, Ross Bros., and Montgomery.)
COMPUTERS and AUTOMATION for May, 1971
Some Suggested Solutions
Much operator verification time can be eliminated by
computer editing (i.e., having the computer verify the
accuracy of the data by a variety of logic checks.) More
realistic scheduling may level out workloads and raise
productivity. In addition, simply measuring operator performance may increase productivity and decrease errors.
New equipment may also offer some further savings and
better control. Many advocates of keypunch replacements
("buffered" keypunch, keyboard-to tape, and keyboard
to-disc-to-tape) promote this equipment as a way to save
from 10 percent to 50 percent of data entry costs.
The accompanying chart of major data entry equipment
characteristics illustrates the general advantages and disadvantages of each category. It does not, however, tell the
whole story. Usually, a careful appraisal of an individual
company's requirements, correlated with a detailed analysis
of critical factors, is needed before a clear picture of the
actual efficiency gains and savings of cost emerges.
Predicting Operator Productivity
Operator productivity is the most difficult, yet the most
important factor to assess.
Experience with the automation of data input and
output has led to the development of a technique for
predicting data preparation times (without physical observation of a job) before actual installation. The technique is
reasonably accurate, and permits a quick judgment of the
19
DATA ENTRY EQUIPMENT - MAJOR CHARACTERISTICS
"0
.=
t)
s::
4. Management control
=
I-<
co C.
A. Are operator productivity and error statistics
easy to obtain?
"0>'
s::
Q)
co~
+-'
en
YES
NO
1. Operator productivity
B. Does the equipment facilitate good, close supervisory control?
A. Is operator limited by machine speeds of duplicating, skipping, and card positioning?
YES
NO
NO
NO
YES
5. Operating costs
LOW
YES
C. Are more than two formats readily available to
an operator?
NO
YES
A. What is the relative unit cost of equipment?
B. Is job setup easy and fast?
NO
NO
NO
NO
IN MOST
CASES
NO
MEDIUM
MEDIml
MEDIUM to
HIGH3
B. What is the cost of supplies?
HIGH
YES
HIGH
LOW
VERY LOW
Co Can subsequent computer processing costs be
reduced?
D. Is it easy and fast to correct detected errors?
NO
YES
YES
NO
YES
E. Can on-line editing reduce the need for verification?
u.
NO
NO
SmlEWHAT
YES
Is a special physical environment required?
NO
NO
\'0
NO
YES
IN SOME
CASES
I\, smm
CASES
E, What are the operating conditions?
2. Error control
A. Is batch balancing an available option?
NO
YES
YES
YES
B. Are edit checks possible (beyond the usual
POOR
POOR
GOOD
GOOD
F. Is the system vulnerable to equipment failure?
NO
NO
SOMEWHAT
YES
field definition controls)?
NO
NO
NO
YES
3. Machine utilization
A. Can one machine both enter and verify data?
NO
IN MOST
CASES
YES
YES
effect on operator productivity of various equipment,
formats, or methods. Preparation times are computed using
pre,determined standards for each individual machine and
each human element required in a data entry job.
For example, consider a data entry job of 3D-character
records in an 80-column card "image" (20 key-entered, 10
duplicated, 50 skipped). Assuming an even mixture of
alphanumeric data, an ideal source document, and an
operator-detected-and-corrected-error rate of 7 per 1,000
keystrokes, the studies showed a productivity factor of .76
for key-tape. This figure means that this job can be
prepared on key-tape in 76 percent of the time it would
take on standard keypunch equipment.
Even greater operator productivity could be realized on
this job during verification because of the ease with which
detected errors can be corrected.
1. With automatic program load option; otherwise,
job setup requires keying in a new program.
2. Most key entry devices can be equipped with
registers to record production and error statistics, but these must be accumulated and
analyzed by hand.
3. From medium to high, depending on the number of
operator stations.
Evaluation of major jobs, using this approach, should
establish the productivity gains and economic benefits that
could be achieved by varying factors.
Some Results
One company, which had to replace its current equipment in order to attain compatibility with a new computer
system, expects to save 4 out of 14 operators by data entry
measurement and control, and another 2 because of improved equipment.
Another company, by applying basic production and
schedule controls, and revising formats, reduced errors to
1/5 of the previous level, increased opera tor productivity
by 25 percent, brought back in-house work (which had
been done by a service bureau), and decreased its work
force - and yet made no change in its key-tape equipment.
o
20
COMPUTERS and AUTOMATION for May, 1971
MAYBE THE COMPUTERS CAN SAVE US AFTER ALL
"If part of the average citizen's feeling of impotence and disillusionment is
caused by a lack of organized and readily-available information, would it not
be possible to put such information at his fingertips with a computer?"
Edward Y ourdon
527 Third St.
Brooklyn, N. Y. 11215
We have known for several years that mortal men are
incapable of managing large cities to the general satisfaction
of their constituents. The 1970 elections have shown us
that this pervasive feeling of disgruntlement has spread to
the state level, and the 1972 elections may very well show
evidence of a "throw the rascals out" mood at the national
level. Perhaps then we will all agree that our society has
grown too complex to be effectively managed by mere
human beings.
Losing Faith
Before this national disenchantment became so strong,
most Americans had what so many generations of men
before us have had: faith. Men must have faith in themselves, in their social order, their leaders, and their government, or we would never have progressed past the point of
feudal kingdoms. That faith usually persists even when the
leaders of a country are slightly corrupt and inefficient, and
even when the people are slightly hungry and unhappy. I t is
usually broken only by catastrophic failings of the system,
failings as dramatic as those that brought on the French
Revolution, the Bolshevik Revolution, or even our own
American Revolution.
Somelfow, we seem to be losing that kind of fai th today,
for reasons that are not entirely clear. Perhaps it is because
poli tical faith, like religious faith, depends on the fai thful
being kept relatively uninformed and unenlightened. Certainly, television and radio have brought us into much
closer contact with those who would lead us. It is possible
that this familiarity has begun to breed a slight degree of
contempt for leaders who, after all, do contradict themselves and who occasionally do use poor grammar.
Perhaps our disenchantment is caused by the fact that
faith seems to work only when surrounded by ritual, by
familiar day-after-day repetitions of the same facts, the
same speeches, the same political ceremonies. Maybe we
have lost faith because Alvin Toffler's "future shock" has
destroyed such comfortable rituals for us.
COMPUTERS and AUTOMATION for May, 1971
What Choices Do We Have?
Given this state of affairs, there seem to be several
choices open to us; it seems to me that we must make a
choice, either consciously or by default, by the end of this
decade. The most obvious choice, of course, is to avoid
doing anything at all. We lack a prophet who could tell us
with certainty what will happen if we fail to make some
basic changes in the form and structure of our government;
at the very least, it would appear that we can look forward
to great periods of restlessness and instability.
We have, sadly enough, no prophets; we have only
leaders who persist in telling us "what's right with America". As a result, it seems likely that we will continue
mUddling along as we have been for the last several
centuries.
Running Away
There are more drastic choices: we can attempt to bomb
our outmoded society into oblivion, or we can simply run
away from it. However, despite the wild antics of a few
revolutionaries, I think the vast majority of Americans are
profoundly committed and dedicated to this country. We
were born here; we grew up here, for better or for worse; a
number of us have fought on strange continents for the
honor and the glory of America. We have no place else to
go; we could not escape the influence of this country and
its problems even if we wanted to. Those of us who run
away to Canada or England are, for the most part, unhappy. Despite the way we move restlessly from city to
city, America is, in the very deepest sense of the word, our'
home.
Back to a Rural Life
Still another choice is represented by the hippie communes and the "back to the earth" movement currently in
vogue with the young. While this is a viable alternative for
21
Edward Yourdon is an independent consultant,
lecturer, author, critic, teacher, and student. He
has published a number of articles on time-sharing
and on-line systems, and his book, Design of OnLine Computer Systems, will be published by
Prentice-Hall next spring. He was previously
Director of Research & Development at E.L.I.
Computer Time-Sharing, and has also worked for
General Electric and Digital Equipment Corp.
Mr. Your don has a B.S. in mathematics from MIT,
and is currently working toward his Ph.D. in computer sciences at Brooklyn Poly tech.
those who simply cannot cope with the Establishment, or
for those who can afford to buy a farm in Vermont, it does
not seem likely that the great mass of Americans would be
either willing or able to settle down to the quiet rural life
our forefathers knew. It is difficult to imagine that a rural
economy would be capable of supporting 200 million
Americans in the lavish style to which they have become
accustomed.
A More Responsive, Computerized Government
There is yet another choice, though it seems to arouse
violent feelings of paranoia whenever it is mentioned: we
can attempt to make our government more automatic,
more organized, and more responsive with the use of
computers. At the moment, the vast majority of government computers are nothing more than glorified adding
machines, used to spit out bills and process tax reports. We
have made little or no use of the computer's ability to
organize and retrieve information, information which could
be used to help legislators govern more effectively, and
which might help the average citizen better understand
what is going on in government. There are a multitude of
major changes which could be effected with computers,
including the six which follow.
1. Voter Registration and Vote Processing
Computers have occasionally been used on the local and
state level to help automate both voter registration and the
actual counting of votes. While these efforts have been only
partially successful, there is reason to believe that a national, unified computerized voting system would remove
many of the inequities in our voting process. Since the
computer would be capable of keeping tabs on every
citizen, it would be easier to relax residency requirements
so that everyone could vote in the national elections, if not
in the state and local elections. The voting booths could be
connected directly to a central computer complex, so that
an individual's vote would be registered immediately. This
would make it more difficult for political bosses to rig an
election, though the possibility of fraud would not be
completely eliminated.
22
2. I mproved I nformation for Legislators
A good deal of the actual legislative work at the state
and national levels is done by staff organizations; the
lawmakers themselves are too busy (or so they would have
us believe) to spend much time analyzing and researching
information on legislation currently under consideration.
Even with this staff help, though, the legislator must often
make snap decisions in areas where he is relatively uninformed. A computerized information retrieval system
would make it possible for a Congressman or a Senator to
obtain information on any subject with a minimum of
effort. The same kind of system might be used 'to provide
citizens information on various subjects, as we shall discuss
in detail presently.
3. Optimum Scheduling of Services
At the city and state level, it often seems that things like
garbage collection, snow removal, road repair, bus schedules
and street cleaning are performed on an inflexible, if not
completely random, basis. Computers could be used to
optimize such services on a relatively dynamic basis, so that
as conditions changed, the services could be re-scheduled.
4. More Up-to-date Information
on the State of the Economy
At the moment, there are a number of economic
indicators which give government officials a rough feeling
for the direction in which the economy is moving. In
addition to the fact that these indicators are often contradictory and subject to different interpretations, there is a
problem caused by the delay from the time the economic
phenomenon occurs until it is noticed; and then the delay
from the time it is noticed until something is done about it;
and finally, the delay from the time some action is taken
until the time the effect of that action is felt. The total
delay can easily be as little as six months or as much as two
years, and it can cause a great deal of economic damage.
What we need is a computerized model of our economy,
an idea which has been of fundamental interest to economists for several years. To have any validity, the model
would have to have several thousand inputs, including such
things as the production and capital expenditure figures
COMPUTERS and AUTOMATION for May, 1971
from major corporations, money supply and interest rates
of the major banks, and employment, wages and spending
figures of the American citizen. The output from the model
could help economists and lawmakers review the state of
the economy on a weekly, or even a daily, basis. Instead of
taking six months or a year to react to an economic crisis,
we would be able to take action within a matter of days.
Even more important, a comprehensive model would
allow economists to simulate the effects of various proposed economic activities. The effect of the General Motors
strike could be predicted by the machine, as well as the
effect of a decrease in defense spending.
people - the economists, legislators, and administrators often have a difficult time describing their application in
sufficiently precise detail for the computer people. Fortunately, there has been a growing familiarity with computers in these professional disciplines, and there is some
hope that they will be willing and able to participate in
more ambitious computer projects in the near future.
Computer people, by the same token, are beginning to
specialize in specific applications, and they should eventually be able to converse more intelligently with legislators
and economists.
People Fear Computers
5. More Streamlined Administration of Government
It is conceivable that a large number of clerical tasks
carried on by local, state and Federal administrations could
be eliminated with a computer. Programs like Social Security, unemployment benefits, the processing of marriage
licenses, and so forth, could almost be completely automated; exceptional cases, of course, would continue to be
handled by people.
In the past, it has often been true that the computers
were more expensive than the people they replaced. However, computers do not ask for raises, do not go on strike,
and do not take long vacations. Since the cost of computing
equipment has been decreasing as a result of improved
technology, and since labor costs continue to spiral upwards, it might be wise for many administrators to reexamine the economics of automation.
6. Better Determination and Control
of National and State Priorities
A common complaint in these times of tight money is
that our priorities are wrong - we should be spending more
on urban problems and less on foreign aid, or more for
foreign aid and less for defense. The final decisions must, of
course, be made by people, and the decisions often take
highly political considerations into account. Nevertheless, a
c01)1puter might be able to help in the decision-making
process. The computer could, for example, easily tell a
legislator how many extra schools or hospitals could have
been built with the money being spent in Cambodia; it
could tell how many jobs would be affected if $1 billion
was shifted from defense work to mass transit or pollution
control.
Equally important, a computer might be able to help
control these priorities, once they were determined. It
could point out cases of fraud and embezzlement, as well as
pork-barrel projects and cases of extreme nepotism.
There is one obstacle, however, that will be more
difficult to overcome: the ordinary man in the street is
deathly afraid of computers. To many Americans, the word
"computer" is a reminder of incorrect bills, all-digit telephone numbers, and the indignity of having to use one's
Social Security number as a prime means of identification;
to other more sensitive souls, "computer" evokes memories
of George Orwell's 1984 or Karl Capek's R. U.R.
There is no doubt that a computer can bungle simple
things like bills and invoices. In fact, when it comes to
resolving an incorrect bill, we seem to be finding that a
computer can be more petty, more arbitrary and more
obstinate than any human bureaucrat. On the other hand,
computer technicians are quick to point out that these
problems are rarely, if ever, the fault of the computer per
se; what has happened is that somebody has programmed
the computer in a petty, arbitrary and obstinate way. If
someone took the trouble to program a computer to be
sweet, apologetic and understanding, much of the ill will
toward computers would disappear.
For example, consider the fact that many current computer systems use numbers as a prime means of identification. When dealing with Blue Cross, American Express, or
the local gas and electric company, one must know one's
account number or there is no hope of getting anywhere
with either the computer or its human attendants. Account
numbers are used primarily because it is easier for the
programmers to deal with a well-known 9-digit decimal
number than it is to deal with a variable-length string of
numbers and letters that he would find in a name_ and
address; in addition, the programmer can be sure that the
account number is a unique identification of the customer,
while the name "John Smith" may not be unique. Nevertheless, the programmer could, if he wan ted to make the
computer system a little more palatable to its customers,
dispense with the ubiquitous numbers forever.
Why Hasn't Government Become More Computerized?
Misuse of Computers
Some of these projects have been attempted on an
experimental basis - the city of Wichita Falls, Texas, for
example, is almost completely computerized. However,
there has been very little concerted effort on the part of the
Federal and state governments to move in this direction,
and things are even more primitive at the city level.
Part of the reason for this backwardness is that many of
the projects are difficult to define and specify. Most
computer people know little or nothing of the application
they are attempting to computerize, and they fail to
program the computer for the exceptions that are inevitably present. On the other hand, applications-oriented
Unfortunately, this does not dispel the deeper fear of
computers felt by laymen and scientists alike: the fear that
the computers will eventually "take over" and start running
our lives. If this happens, it will not be the resul t of the
computer having acquired some innate intelligence of its
own - while we can get computers to play a reasonably
good game of tic-tac-toe and checkers, we computer people
have all but given up hope, for the present time at least, of
building a truly "intelligerlt" computer like HAL in the
movie 2001. The really important danger, as Norbert
Wiener has pointed out in books like God and Golem and
The Human Use of Human Beings, is that an unscrupulous
COMPUTERS and AUTOMATION for May, 1971
23
leader can use a powerful computer to help subjugate his
people, or that a thoughtless leader might abdicate some of
his decision-making powers to a computer. The appearance
of com pu terized war games, computerized military
strategy-making systems like WIMMIX, computerized defense systems like SAGE, and the growth of computerized
surveillance files certainly lend credence to these fears.
It would be absurd to minimize the dangers to this kind
of misuse of computers - dangers that seem potentially far
greater than those posed by the computers that generate
incorrect bills. Since it is ultimately people that misuse the
power of a computer, just as it is people that misuse atomic
energy, a great deal more attention should be given to
systems of human checks and balances to ensure that the
rights and privileges of American citizens (and citizens of
the rest of the world, for that matter) are not being
endangered by computers.
Can We Have Faith in Computers?
It seems, then, that computers could bring about a
tremendous improvement in various phases of government
... if one has faith: faith that the computers will work
properly, faith that they will not be as petty and obstinate
as many of the current computer systems, faith that they
will not be misused by scheming politicians or over-zealous
bureaucrats. We seem to have come full circle, first indicating that men had lost faith in their human leaders, and
now suggesting that things will be better if they have faith
in a cold-blooded mechanical computing machine.
In the long run, the advantages of computers will
hopefully become self-evident. If, twenty years from now,
people become generally aware that it is a computer that
gets the garbage picked up on schedule; a computer that
makes the telephone work properly; and a computer that
keeps unemployment at a minimum, then they may gradually begin to feel a little more benevolent towards the
machines.
Opinions: A Mixture of Facts and Faith
In the meantime, the garbage iSIl't being picked up, the
telephone doeSI1 't work, and unemployment is by no means
at a minimum level. To make matters worse, even the
pitifully primitive computer systems that currently exist
don't work half the time. As a result, many of us are
perpetually disgruntled, and do not feel kindly towards
politicians, computers or the government. Our opinions on
these subjects are a curious mixture of facts and faith, and
the proportion of the two seems to differ from one
generation to the next, from one neighborhood to the next,
and from one ethnic group to the next. It is important to
realize, I think, that our opinions and our views of the
world are a function of our environment. Most of us have
certain social and political attitudes formed and influenced
by our parents, our friends, by the type of education we
received, by the type of work we do, and by the newspapers, movies, and television shows that we happen to
watch.
One of the problems in forming an opinion is that there
is simply too much information available on any particular
subject. In order to form an intelligent opinion on Vietnam,
for example, one is faced with the difficult task of reading
dozens of books, thousands of newspaper articles, reprints
of Congressional hearings and investigations, as well as the
24
speeches and statements by officials of the government and
military. There is simply too much to read, too much to
learn, and too much to keep up with. Our natural instinct,
in many cases, is to block it out, or avoid it. Depending on
our inclinations, we read selected articles from the New
York Times and avoid the New York Daily News; we read
Time magazine, but not Ramparts; we listen to speeches by
Barry Goldwater, but turn off the television set whenever
Eugene McCarthy's face comes into view; we expeditiously
avoid books that run counter to our opinion on the subject.
On the other haild, there are times when desperatelyneeded information is not available to the average American. Listening to a debate between any two political
candidates, for example, can be a highly frustrating experience - each accuses the other of having wrongly opposed
or supported critical legislation, and it is extremely difficult, given the resources and the patience of the average
citizen, to detect who is telling the truth. The "truth", such
as it is, may be scattered through various official documents
and reports, or it may be withheld from the public for
reasons of national security and/or political expedience. It
is highly ironic, given the nature of our national malaise, to
hear leaders implore us to "have faith" that their programs
will work out well.
Everyone Asks "Why?"
We are faced with a serious dilemma. Any manager or
leader will point out that he cannot do his job properly if
he constantly has to report every detail of his decisionmaking processes to his subordinates; this maxim becomes
more and more true the higher one goes in management of
any kind. On the other hand, the citizens of this country
are faced with such staggering problems that they are no
longer willing to submit meekly to the decisions of men
who, as they have seen on their television screens, are often
not much wiser than they. Everyone now wants to know
"why": why haven't we gotten out of Vietnam, or conversely, why haven't we invaded North Vietnam? Why are
we spending so much money on this program or that
program? Since adequate answers to these questions are not
forthcoming from our leaders, and since sufficient information is not available for the average citizen to answer his
own questions, we have seen a growing disillusionment with
all politicians.
A Computer in the Hands of the People
One way of improving the situation would be to put a
computer in the hands of the people. If, as we have
suggested above, part of the average citizen's feeling of
impotence is caused by a lack of organized and readilyavailable information, would it not be possible to put such
information at his fingertips with a computer?
I believe that it would be feasible, both economically
and technically, to create a National Information Bureau,
whose sole purpose would be to provide information to any
citizen on any issue.
Creating a National Information Bureau
An example will illustrate the possibilities of such a
system. Suppose a local citizen's group wanted more information on welfare, so that it could form an intelligent
opinion of local political candidates. Not knowing where to
begin, it might ask the National Information Bureau what
COMPUTERS and AUTOMATION for May, 1971
information existed on welfare. The Bureau might respond
that its files on welfare are broken into six categories:
1.
2.
3.
4.
5.
Welfare legislation
Welfare statistics
History of welfare in the United States
Sociological and psychological effects of welfare
Attitudes towards welfare - speeches, interviews,
etc.
6. Bibliography
Each of these categories could, of course, be broken
down further. Thus, the section on "legislation" might be
broken down as follows:
1. Legislation
a. Federal legislation
1. Currently pending legislation
2. Recently enacted legislation
b. State legislation
1. Alabama
2. Alaska
•
•
•
c.
50. Wyoming
Legislation passed by major cities
1. New York
2. Boston
etc.
The local citizen's group could then request summaries
of books, copies of articles, speeches and so forth. Similar
information could be maintained on such subjects as Vietnam, defense spending, the economy, and crime. Information would be available in as much or as little detail as
desired.
The major purpose of the National Information Bureau
would, of course, be to serve as a central source of
information on any subject of reasonable interest. However,
it would also serve to illustrate the inconsistencies and the
contradictions that exist in areas like the Vietnam war. If a
political figure made one speech in the North and another
contradictory one in the South, it would become evident in
the files of the National Information Bureau; if his voting
record in Congress was at odds with his public speeches to
his constituents, it would also be recorded by the Bureau. If
a politician quoted statistics or made charges that were
contradicted by other reputable sources, that, too, would
show up in the files.
country has a right to a decent amount of food, clothing
and shelter, and the conservative feeling that everyone
should work hard enough to be self-supporting. In other
situations, the Bureau might show that there are five or six
sides to an issue, each of which has its own combination of
facts and faith.
The National Information Bureau would essentially be a
large computer system. It would receive newspaper articles
from every major newspaper in the country; speeches by all
major public figures; books; polls, news analyses by television and radio commentators, and so forth. All of this
would be filed, categorized and summarized automatically.
The source material' might be kept on microfilm; summaries, analyses and indexes could be kept on faster forms
of storage. For those who merely wanted to "browse",
information could be displayed on devices called CRTs,
which look like television screens. Copies of source documents could be made on high-speed printers or microfilm
reproduction equipment.
In computer parlance, such a system is known as an
"information retrieval system". As long as the range of
information is fairly well-defined, and the relationships
between different pieces of information not too complex,
information retrieval systems are certainly within the capabilities of current technology. A number of business organizations use information retrieval systems to extract information about their employees or about sales, production or
inventory. Scientists often use information retrieval systems
to find literature on a particular subject of interest. Even
the government data files which pose such a potential
threat to our privacy are, for the most part, information
retrieval systems. The National Information Bureau would
simply be an information retrieval system designed to
handle a different kind of information for a different
clientele.
Financing the National Information Bureau
Since computers are so expensive, financing such a
system might well be a problem. It would be desirable to
avoid government financing, since that would pose a number of thorny problems. On the other hand, it is not at all
clear that the National Information Bureau could be selfsupporting. To do so, it would have to charge its customers
for the information it provided, and this would certainly
discourage both the poor and many of the middle-class
Americans who desperately need it. The only approach that
seems viable, at the moment, is to provide financing from
an independent, non-profit foundation.
False Data
Helping to Distinguish Between Fact and Faith
The National Information Bureau could also help people
distinguish between matters of fact and matters of political
faith. It might become evident, for example, that despite all
the statements and speeches emanating from the White
House, Richard Nixon's attitude toward youth is still
summed up by his· comment to Theodore White: "They
were given too much, too easily; and this weakened them."
In addition to showing what the public figures really believe
in, the files of the Bureau could also show what "faiths" are
involved in the major issues of the day. It might show, for
example, that the question of welfare really boils down to
an emotional argument between the liberal and conservative
attitudes: the liberal feeling that everyone in a civilized
COMPUTERS and AUTOMATION for May, 1971
There arc other potential problems that should be explored before any money is invested in such an ambitious
undertaking. For one thing, it is quite possible, if not
almost certain, that false data would be supplied to the
system at various times. Economic or military figures which
would prove embarrassing would probably be "adjusted"
before being released to the public and to the files of the
National Information Bureau. This, of course, is already
being done, but the exposure that would be provided by
the Bureau would make it even more necessary. It might
also make politicians attempt to withhold more information from the public.
This would not cause any great harm to the information
retrieval system unless it took the form of a complete
25
national conspiracy. There are still some politicians in this
country with opposing viewpoints, and there is still some
free flow of information; since all available information
would be digested by the Bureau, the deceptions or inconsistencies of anyone politician would soon become apparent. However, if the entire United States Government
undertook a concerted effort to hoodwink the American
people, it is conceivable that they could engage in such a
massive propaganda campaign that even the National Information Bureau would be fooled.
The Bureau Might Be Biased
There is also the possibility that the Bureau itself might
be biased. If the computer were programmed by a rabid
segregationist, for example, the resulting system might
reflect that bias. Since the National Information Bureau
would be dealing with the categorization, the summarization and, to some extent, the analysis of information, it
would be easy for even the most subtle personal prejudice
to work its way into the computer. Hopefully, this problem
could be resolved by subjecting the system to a constant
scrutiny by people of varying political and philosophical
attitudes.
Criticism and Pressure
I t is almost certain that the system would be subjected
to extreme criticism by any individual or group that felt it
was being unfairly portrayed by the Bureau. Financial
and/or political pressure could certainly be brought to bear
on the system by lobbies, companies, individual politicians
or even the entire government. If this happens, and if the
pressure is strong enough to shut down the National
Information Bureau, then we will indeed be in as much
trouble as the young revolutionaries say we are.
NUMBLES
NUMBER PUZZLES FOR NIMBLE MINDS
-AND COMPUTERS
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 0 to 9. A digit may be represented by more
than one letter. The second message, which is expressed in
numerical digits, is to be translated (using the same key)
into letters so that it may be read; but the spelling uses
puns or is otherwise irregular, to discourage cryptanalytic
methods of deciphering.
We invite our readers to send us solutions, together with
hunun programs or computer programs which will produce
the solutions. This month's Numble was contributed by:
Stuart Freudberg
Newton High School
Newton, Mass.
NUMBLE 715
S ELF
+ T R U S T
C 0 F F E
The Greatest Danger: Lack of Interest
The greatest danger of all is that nobody will be
interested in such a system. No matter how available the
information is, there will certainly be some people that will
be too lazy or too uninterested to obtain it. There may also
be a number of people who will not want to have anything
to do with the Bureau because it only seems to tell them
bad things. The "it's-about-time-we-heard-what'sright-with-America" philosophy might find it very difficult
to cope with a National Information Bureau that refused to
sugar-coat the material it collected.
It is not really clear, then, whether such a system would
work or would have any value. It is clear, though, that
computers are here to stay; technology is here to stay; the
information explosion is here to stay. American society will
continue to become more complex, more confusing, and
more technological in nature. No one seriously expects
Americans to move back to the farms ell masse in order to
solve the ecology problem. Similarly, no one really expects
the telephone company to give up its all-digit dialing
system, or the government to give up on its attempts to
reduce everyone to a social security number.
We have already spawned the monster; now all we can
do is attempt to control its growth and its appetite. We can
try to control the way technology is used, so that it
becomes easier, not more difficult, for people to cope with
the complexities of our age. The National Information
Bureau might well be a first step in that direction.
D
26
H=L
+ISTHE
+ 0 FIE F I
56748
43974
81542
99344
Solution to Number 714
In Numble 714 in the March issue, the digits 0 through 9
are represented by letters as follows:
T=O
M,N= 1
A,U=2
S=3
L=4
F,G = 5
I,Y = 6
0= 7
E=8
D=9
The message is: Folly does not see its magnitude.
Our thanks to the following individuals for submitting
their solutions - to Numble 713: Gordon Bruno, Cliffside
Park, N.J.; A. Sanford Brown, Dallas, Texas; T. P. Finn,
Indianapolis, Ind.; John H. MacMullen, Eden Prairie, Minn.;
Richard Marsh, Washington, D.C.; G. P. Petersen, St. Petersburg, Fla.; and Robert R. Weden, Edina, Minn. - to
Numble 712: Twite S. Emerick, Harrisburg, Pa., and Vincent K. Roach, New York, N.Y.
COMPUTERS and AUTOMATION for May, 1971
THE ASSASSINATION OF PRESIDENT KENNEDY:
The Spatial Chart of Events in Dealey Plaza
In the May 1970 issue, "Computers and Automation"
published a 32 paqe feature article by Richard E.
Spraque, entitled "The Assassination of President
Kennedy: the Application of Computers to the Photoqraphic Evidence".
In this article, Richard E. Spraque, President,
Personal Data Services, Hartsdale, N.Y., stated
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; and indicates convincinqly
that there were at least four ounmen firina from
four locations, none of whom was Oswald; and that
the conspiracy to kill Kennedy involved over 50
persons (of whom several are identified in the article) •
This article contained a spatial chart of the
events in Dealey Plaza, Dallas, Texas, at the time
of the assassination of President Kennedy, showing
many details. The chart published here is a revision and contains corrections and some additional
information: the scale in feet has been corrected;
certain cars have been properly titled; and four
or five names of persons or locations have been
added or corrected. C&A will publish additional
revisions and corrections when known.
1. Areas on the Chart. The spatial chart is
divided into square areas 40 feet by 40 feet, each
labeled by a letter A to K from top to bottom (I is
omitted) and a number 1 to 15 from left to right.
2. Person Names. The name of a person in the following index refers to the location of such person.
3. Numbers. A number folowing the name of a person refers to the film (or roll) number of a still
photograph, or the frame number of a movie sequence
taken by such person. Numbers preceded by Z refer
to frame numbers of the color movie film taken by
Abraham Zap ruder standing in area D 7.
4. Motorcade. The leading portion of the motorcade including President J.F. Kennedy is shown
diagrammatically 7 to 8 seconds before the first
shot. The motorcade is arranged in sequence along
Houston St. and Main St. All these cars, of course,
were moving and therefore occupied different positions at later times. The only successive locations
shown for later positions of the motorcade are the
locations of President Kennedy's head in the lead
car. These head locations are marked by a sequence
of dots along Elm St. These locations were carefully
determined by the FBI; they are identified by the
frame numbers of the color movie film taken by Abraham Zapruder; these locations were used by the
Warren Commission in their investigation. Successive
frames were 1/18 second apart.
5. Note. The information shown in the diagram of
the motorcade (C-J 14, J 15) is not repeated in this
index.
Much of the geographic information (such as locations of trees, white traffic lines, yellow marks
on curbs, etc.) is not repeated in this index.
INDEX AND GUIDE TO LOCATIONS
Location
Person or Object
Altgens 2, 3
Altgens 6 (at Z 255)
Altgens 7 (at Z 346)
~
F 8
G 8
'''Babushka Lady" (who took
an entire movie of the
motorcade from the opposi te side from Zapruder)
Bell I, 2
Bell 3
Betzner
Betzner
Betzner 3
Bond, 1 to 3
Bond, 4 to 9
Brchm(s)
Brenna n
bullet mark(s) on curb
Commerce St.
compass directions
Connally, Gov. J. B.,
back shot (at Z 238)
Hester(s)
Hill, Jean
Holland
Houston St.
Hudson and two friends
Hughes I, 2
Hughes 3
Hughes 5
H
H
E
C
C
G
G
F
D
J 3, H
11
11
13
13
12
13
11
9
13
6
K 1-6
J 8
E 9
Da lla s County Crimi na 1
Courts Bldg.
Dallas County Records Bldg.
Dallas County Sheriff's
Office
Dal-Tex Bldg.
Dillard 1
Dorman, Mrs.
Elm St.
Elm St. extension
former Dallas County
Courts Bldg.
garage bldg. next to
TSBD
gra ssy knoll
F 8
F-H 15
C-E 15
H 15
A 15
C 13
A 11
H 1 to B 15
A-B 6-12
off map, see K 15
A 8-10
G 4 to C 10
C 7
F
9
H 2
A-K 13-14
F
6
14
K 12
K 10
K
COMPUTERS and AUTOMATION for May, 1971
Person or Object
Kennedy behind "the"
oak tree, from Z 161
to Z 207
Kennedy at:
Z 133
Z 161
Z 189 (throat shot)
Z 226 (back shot)
Z 238
Z 255
Z 285
Z 312 (head shot)
Z 313 (2nd head shot)
Z 346
Z 400
Z 433
Z 465
Z 485
knoll, grassy
Main S1.
Man tf- •••
these were
men whose names have
not been determined
Man tf-l, source of 1st
shot at Z 189
Man tf-2
Man tf-3
Man tf-4
Man tf-5, source of 6th shot
at Z 313
Man tf-6, off map in Dal-Tex
Bldg, source of 2nd shot
at Z 226 and 4th shot
at Z 285
Man tf-7
Man tf-8, source of 3rd shot
at Z 238 and 5th shot at
Z 312
Man wi th Umbrella
Martin 0
Martin 1
Martin 2
material picked up by
-Walthers and others
Moorman
Muchmore 1
Muchmore 2
Location
D 10-11
C 11
D 11
D 10
E 9
Person or Object
Newman(s)
Nix I, 2a
Nix 2B
Oswald's alleged window
parking area and railroad
yard
puff of smoke (shown in 9
photos)
railroad yard and parking
area
Location
-Ef3
K 12
J 10
A 13
B-E 4-5
E 7
B-E 4-5
F 8
G
G
H
H
H
to C
6
5
3
1
9
scale in feet
shots, sources of:
1st
2nd
3rd
4th
5th
6th
K 8-9
E 5
A 15
A 11
A 15
A 11
D 6
J 1-15
D 5
E
E
E
D
7
7
7
6
see A 15
D 8
A 11
D
H
G
C
H
8
13
13
12
6
F 9
G 13
F 11
Si tzma n
smoke, puff of (sho~n in
9 photos)
Stcmmons -·Freeway sign
D 7
E 7
Tague (~ho was hi t in the
face by a fTagment of
a shot)
Texas School Book Depository Bldg. at 411 Elm
St. (TSBD)
TSBD, 6th floor easternmost
window, from ~hich Warren
Commi ssi on a 11eged Lee
Oswald fired 3 shots
Towner, J. 1
Towner, T. 1
Tcwner, T. 2
K 3
"Umbrella" Man
Willis 5, 6
Zapruder (location of
Abraham Zapruder during
the entire color movie
··,hich he took)
D 8
A 10-13
A 13
C 13
C 13
D
12
D 8
D 11
D 7
(See chart on next two pages)
27
of
C.~AR..T
.sPATIAL
NOR..TWE....R-N
WALF-
of
:PLALA
DEALEV
DALLA.s, TE..XA:;'
!>h..oWln..~
J
0
oh.oto91"'0-Ph..e..I"!>. posd'"\on...~
Ken.rvuilf~
or
o6Je.L~ {.
sUr"i'oun-d.\..nC} \1...~)O pm I 'Z.1.. Nov'. \~b)
h..n. f-.
uuU!.
h.R.o.d..
LE~ E."'·,u:~
•
f.cJ.rn.
~,-ud(U'
Z-I)3
fv-cun~
locArlnC) JFK.:£ ~
n.um..b.t.l- 4- dot
probo..bk Lu-u..- o~ fU-e.. to h.i.t
mUL u)h.o~(. n.o..rne..S h.o.YQ.. nOlo.!.
y~t be.e,\"1 d.e-Tu-mlne..d..
I: people..
:la.m.p post
...fOb.:: storm !>e.u) u- \n LU:
L)m: '}dlow ,,..ta.rk. on. c..u.\"6
@
old. wood.
ti.u u:,c.cl. cu -7.
bLlm.~.
c..,rb
-:::::::::-.:J: ~e.. cii.o..t)'-a..rnma,jic.: not to suJe.
I
91"'0......
:)f'r~wt..
pbto. i.rttu-~'~ILTe..r
R f> c.ut\~· rUt..a.l-c.hu-
p.....e.pOS't.d
up
wi.ndoUJ
~;
R,..E..
CU'"Ul.
~ie.p.$ =
on
+.a.
Secone!. :p ....U"\t~~ , M
CU} 1~71
~
tunn&! ulo.l\
f':a.-
.-
\
\
rrl'
\
~olLo..ru!· \
on
OVt.rpIUS
ELM
~~
~.=
___.
\
!)t
\
~
--~·----':---=--:-~':::=--======tiiiiiiiiiii\Iiii=~~~=-"--~c..==~T~'Pl.E...
J
~-~ f?:<
,~-j~,:=~-
'. r·
101 '.
<1/' I
=~=~'=========~===r
~
UNO!.,,-PA,-SS
MAIN
5t
LQJ
..
CO M M E R..C E....
~ ..
"
'-.
EL M
2
o
r-
tf'/
E.LM ':.t
:l
o
:r
• DillArcl.I(Ut -P(,.t.~~/('a.rnu-o.. car -# 3
4
11.:~I)
DALL....'5
c.ouNTY
RECCIZD!a
In..l)~
i
•
Be:tirw-I
3
I
JMortin.
1
DALLAS
COUIJTY
/".
C12JMI~
;.~.
COWtT$
':: '
2
'i1
t-JOI2..T"- : "-
&n.cll.l..
, . , ..........-.r
eBt..U.l z.
..... ~ \;:j'\:>r'I"
•
' .. '
I'
5U:IG
;
,j' ',:' " ; __
~: ':' ..
~.
0·· .•
'
•
t\-~"~F
D~.ijGS"~
co.r"3
60
,",,0
I>e..~ .?!
h
~G4.Ie.
."n.,,!
in.
feet
80
I
Weo.ver
PLAYING GAMES WITH A COMPUTER
James R. Palmer
Member of the Technical Staff
Space and Communications Group
Hughes Aircraft Company
Los Angeles, Calif. 90009
Playing games can be fun -- and educational.
Anyone who has played the game Monopoly has probably learned a lesson about undercapitalization and
a liquidity crunch.
The advent of the computer offered a new vehicle
for these lessons: the management game. Overnight a science of computerized education sprang
forth. Statistical simulation was employed to
develop some very elaborate and sophisticated game
models. More recently computerized education has
turned toward the concept of programmed learning as
a vehicle for instruction.
The El Segundo Division of Hughes Aircraft Company was recently faced with a need for utilizing
some of these techniques. The company was developing a detailed educational course on production
control. It was felt that some type of game would
be useful at the end of the course to summarize the
basic principles. A survey of games and simulations
then available uncovered no precisely appropriate
packaged product. Hence authorization was given
for a project to develop an educational game for
this application.
The
D~velopment
Project
From the beginning, the project was oriented
towards a very clear objective: to design a game
that illustrates the basic principles of production
control. But coupled to this objective were three
constraints.
First, the development and operation of the game
had to be inexpensive.
Second, the qame had to be simple.
Third, the game had to prove true-to-life realism.
These constraints were imposed by the particular
application for which the game was designed.
Three educational tools were selected for use
in the game. These were the computer, statistical simulation, and programmed learning. Tne
computer was selected because of its speed, impartiality, and accuracy. To contend with the high
costs attached to its use, a timesharing system was
selected and special emphasis was placed on simplified game design.
cipants toward,considering a set of fundamental
questions that the subject (production control) is
designed to answer. That is, a production control
system is designed to answer the questions:
What is being manufactured?
How is it being manufactured?
How many are being manufactured?
When are they being manufactured?
Therefore, the game was structured to force the
student to repeatedly answer these questions as
the simulation proqressed. Included was a system
of response checking to flag any illogical responses by the student, present an analysis of errors,
and offer a chance for resubmission. This is
similar to standard programmed learning techniques
except that the questions were made cyclically repetitive, and the amount of conditional branching
was sharply curtailed.
The Results
The game that resulted from this project is called
DEFOG, or Deterministic Factory Operation Game.
It is designed around a fictitious product having
five parts. To conform to the "true-to life realism" constraint, a complete set of drawings and cost
and technical specifications for the parts are included in the game. Several days before the simulation
is to be run the participants (who may be separated
into teams) are given an instruction manual for the
game. They are told that for the game they have
been assigned responsibility for meeting a specified
delivery schedule on a fixed price contract. (Penalty costs are established for missed schedules.)
Also, they are given a set of general instructions
for pre-game study and planning.
The actual simulation proceeds in a month-by-month
manner. Each simulation month starts by asking the
participant which parts he wants to purchase, fabcate and assemble during that month. He is then
asked for his order quantities in each category.
His responses are checked to make certain he hasn't
tried to purchase something that can't be purchased,
assemble something he doesn't have parts for, etc.
At the end of each simulated month, a breakdown
of the participant's inventory and manufacturing
cost is presented. At the end of the game, the
participant's profit on the contract is calculated.
Analysis
Statistical simulation was selected as a tool
because of the atmosphere of challenge and realism
that it offers. But here a rather novel approach
was taken. Because of the constraints on game
simplicity and cost, ~he-introduction of any random factor game elements was avoided. That is,
the game was written in mathematically deterministic format. The goal was to keep the game elements simple enough to be understood but complex
enough so the solution is not immediately obvious.
DEFOG has ,been successfully employed in several
applications. In addition to its use in four classes
of the company's production control course, the
game was presented at the recent 1971 Systems Engineering Conference (SENCO 71) in Phoenix, Arizona.
Post-game informal discussions by the participants
have indicated that they did in fact develop an increased understanding of certain principles of
production control. Comments like "inventory costs
killed us", and "we found a way to avoid assembly
overtime", indicated an understanding of various
cost trade-offs. So it seems that the objective of
the game was met.
The third educational tool selected, programmed
learning, also had to be modified for this application. The objective of the game was to illustrate principles, not to teach specific information.
Hence the approach taken was to guide the parti-
Furthermore, the game does appear to remain within the constraints which governed its design. Total
development cost was under $1500. Labor was by far
the largest cost element. Development time was one
month. It costs less than $9 per team to play the
30
COMPUTERS and AUTOMATION for May, 1971
game. This mayor may not comply with the constraint on being "inexpensive", but the company
feels that DEFOG will be a very profitable investment over the years. With regard to the constraint
on game simplicity, success is more obvious. Of the
four teams that played the game on its premier opening occasions two teams discovered the optimum solution. The other two teams managed a profii on the
simulated contract. Finally, the sense of realism
in the game seems adequate. The first participants
were operating manufacturing men; they were able to
readily identify with the situation and the problem.
Some Generalizations
If the explicitly expressed satisfaction of
DEFOG's users is a legitimate measure of success,
then DEFOG is a success.
But perhaps the most significant success is not
DEFOG itself but rather the teChnique used to
develop it. This technique can be summarized in
fi ve steps:
(1) Determine a specific need or application~
because it forms a particular part of a
particular production control course being
given to a particular group of individuals. In a different application DEFOG
(or ~ game) might be irrelevant and
worthless.
(2) Once it is decided that there is a valid
application for a game, determine the precise objective of the game. A game designed to teach specific facts should be
quite different from a game designed to
teach general principles.
(3) Establish the constraints to which the
design must conform. Most important is
the financial constraint. Other constraints will stem from the particular
application.
(4) Determine what tools are available, and
which of those are most appropriate.
(5) Design the game to meet the objective
within the constraints using the selected tools.
Acknowledgements
The author thanks Dr. Harry W. Steinhoff, Jr.,
of Hughes Aircraft Company and Dr. Alan Rowe of
the University of Southern California for their
generous contributions to the success of this project.
ESSENTIAL COMPUTER CONCEPTS FOR TOP MANAGEMENT
1. From Frank J. Gabriel
6740 Old York Road
Philadelphia, PA 19126
In your extraordinarily fine, perhaps the finest
magazine on computer technology developments,
"Computers and Automation", in the October 1970
issue, in your excellent edi torial "What Top Management Should Know About Computers", you stated
that you estimate that a member of top management
would need to know and understand some 60 to 100
important terms and some 40 to 80 propOSitions
about computers.
COMPUTERS and AUTOMATION for May, 1971
May I respectfully suggest, without imposing
on your extremely busy time, that if it is in
any way at"all possible, you Sir should graciously
make that compilation and publish it forthwith
(as it is very much and urgently needed), either
in your magazine or in a separate booklet; and
that I woul~ consider myself privileged to receive
it either on a complementary basis or paying for
it, which I would do gladly.
2. From the Editor
In an editorial in "Computers and Automation"
for October 1970 I said:
I would estimate that a member of top management would need to know and understand some
60 to 100 important terms, and some 40 to 80
important propositions about computers. Also
he should have a short (3 to 5 "day) course
(of good quality) about computers. And the
course should include interactive contact with
a computer.
As a result of Mr. Gabriel's request, I have put
together a beqinning, a preliminary list of slightly over 100 important terms and topics, as candidate terms for top management to understand.
(Definitions and propositions will be determined
later.)
1. General Concepts
computer / data processor
digital computer vs. analog computer
desk calculator vs. slide rule
electronic data processing / automatic data
processing / EDP / ADP
size / cost / speed / capacity / flexibility /
reliabi li ty
communication / computing
2. Computer
definition of / history of / future of /
programming of
computerized system / systems analysis
applications: what a computer can do easily
vs. what a computer can do with difficulty
vs. what a computer cannot do at all
final assembly principle vs. principle of
successive models
stored program computer vs. externally programmed computer
input / output / storage / central processor /
registers or locations / core
external storage: punched cards / magnetic
tape / magnetic discs / punched paper tape
modem / console / terminal
3. Programming
hardware vs. software
machine instruction / machine coding / machine
word
31
memory reference
8. Computer Department
operation code
organization of department
ASCII (pronounced "askee", and meaning American Standard Code for Information Interchange)
open shop vs. closed shop
batch processing vs. time-shared vs.
access vs. entire machine
mote
conditional transfer
operator vs. programmer vs. systems analyst
microprogramming
batch operations vs. real-time operation
pseudo operation code
tu rn-a round-time
standard coding form
4. Programming in General
programming languages
information / data
numbers / digits / characters / truth values /
words / parameters / variables
It seems to me that this is a~most the total
list of essential computer concepts and topics for
top management. This list may even err a little on
the side of fullness. However, a member of fop
management needs to "know his way around" so as
never to be fooled or deceived.
Comments from readers are invited.
mathematics vs. logic vs. programming
algorithms or calculating rules
accuracy vs. precision
problems vs. solutions
constants / expressions / statements / loops /
variables / declarations / transfer instructions / conditional transfer instructions /
subrouti ne calls
octal system vs. binary system vs. decimal
system
fixed point operation vs •. floating point operation
binary arithmetic
simulation / models
5. Programming Languages
machine language
symbolic language
(Based on a report in "Computerworld", March 24,
1971)
A court in Oakland, Calif., has awarded $290,000
to an insurance broker Paul F. Roemer, Jr., plaintiff1in a precedent~setting case in which libel
damages were filed against Retail Credit Co. This
company is the largest dossier holder in private
business, and has some 48 million dossiers on file.
It deals largely with private business but also
cooperates with government agencies.
The charge alleged that damaging information was
supplied through a former business associate who
had had disagreements with Roemer.
The credit company's report on Roemer said that
there was a question of his honesty over misuse of
funds. The conclusion of the report was that Roemer
"is by no means recommended."
("DETAILS" -continued from page 7)
assembly language
-------(may be copied on any piece of paper)-------
COBOL / FORTRAN / BASIC
To: Computers and Automation, Dept. E
815 Washington St. w Newtonville, Mass. 02160
problem-oriented vs. machine-oriented programming languages
6. Programming Operations
writing / debugging / documenting / flow
chart / assembling / compiling / loading /
operating / testing / storinq / dumping /
tracing / debugging / de~k checking
amendment, correction, modification
7. Systems Analysis and Synthesis
observation / interviewing / investigation /
typical examples of results / testing of
typical examples on a computer / bench-mark
problems / feasibility analysis / changeover
/ continued modification and adaptation
32
$290,000 AWARDED IN LIBEL
DAMAGES TO AN INSURANCE
BROKER SUING RETAIL CREDIT CO.
Yes, r am interested in the offer made in the
editorial by E. C. Berkeley on page 6 of the
May 1971 issue.
r am an unemployed computer professional.
r am a subscriber to "Computers and Automation".
r am seriously interested in trying to develop
a small business of my owno
r enclose a statement about my education, background, aptitudes, resour6es. interests, purposes, and any prior experience r have had
as a self-employed person.
Please send me the list of the several dozen
products and services for which you have
observed a need, and some further remarks
and guidelines on the subject of operating
one's own business.
My name and address are attached.
COMPUTERS and AUTOMATION for May, 1971
RESTORATION OF SERVICE AFTER EARTHQUAKE
you offered a copy of a list of books concerning relationships to reality.
D.L. Talley
Redcor Corporation
21200 Victory Blvd.
Woodland Hills, Calif. 91364
I believe I sent a Readers Service Card (with 2
circled), as instructed, to obtain a copy of the
list. To my knowledge, I have received no communication.
Tabulating Consultants, Inc., a major computer
service bureau in Burbank, California, faced a severe financial loss on- Tuesday February 9th. Their
company is located close to the epicenter of the
major earthquake which struck Southern California
at 6:01 am that day. Tabulating Consultants runs
two shifts a day on their 16 terminal KeyLogic system.
If this listing of books is still available, I
would still be interested in obtaining a copy, if
you please. I assume the copy would be available
essentially free of charge.
Jack Moore, one of Redcor Corporation's customer service engineers, reported to Tabulating Consultants when they opened at 7 am. Power was off for
over an hour. Moore assisted in clearing debris.
When power was restored, he ran complete preventive
diagnostic routines. The system was re-initialized
and in a short time turned over to Jeri Borella,
Data Transcription Supervisor for Tabulating Consultants.
Richard Kurzenknabe, Tabulating Consultants'
president, said "I couldn't be more pleased with
the Redcor initiative and assistance in getting
us back into business. Redcor anticipated the
problems which might have resulted from the earthquak
and the resulting power failure. Their response
allowed us to get back on the air with the minimum
amount of lost revenue. You couldn't ask for better service than we received during this emergency."
UNINTERRUPTIBLE POWER SUPPLY PROTECTED
NASA COMPUTER DURING EARTHQUAKE
Robert R. Bentley, News Bureau
General Electric
Bala Cynwyd, PA 19004
When rocked by one of California's strongest
earthquakes on Feb. 9, the computer complex at NASA's
Jet Propulsion Laboratory in California still functioned flawlessly, providing an important link in
returning Apollo 14 astronauts to earth.
The sixty million dollar complex of computers was
serving as a back up monitoring center for that manned flight, as well as receiving data from deep-space
probes, when the earthquake broke incoming power
lines.
An uninterruptible power supply provided continuous power, helping assure safe splashdown for the
astronauts, and save irreplaceable information from
deep-space. Even milli-second interruptions of computer power can cause misinterpreted data or loss of
stored information.
JPL's uninterruptible power system was developed
by the Custom Power Equipmen~ Departm~n~ of General
Electric and it smooths out IrregularItIes and pr?vides battery power during major power breaks untIl
emergency generators can take over and supply power.
liTO HELP LIBERATE ONE'S MIND FROM NEWSPEAK"-COMMENT
Thank you for your assistance.
2. FrQm the Editor
The list was referred to in March with the following description:
I have put together a list of some dozen
books that I think are useful to help liberate one's mind from the propaganda and
Newspeak of today.
Because over 300 readers of "Computers and Automation" asked for the list, we decided to publish it
with comments. In the June 1970 issue it was published with full citations and comments, under the
title "To Help Liberate One's Mind from Newspeak",
on pages 9, 10, and 13. For convenience, the books
referred to there are here listed again, but for
more information please see the June issue.
1.
1.
2.
3.
4.
5.
6.
In General
Bertrand Russell / Sceptical Essays
Rudolf Flesch / The Art of Clear Thinking
Winston W. Little, W. Harold Wilson, and W. Edgar Moore / Applied Logic
Monroe C. Beardsley / Thinking Straight: Princi les of Reasonin for Readers and Writers
George A. Miller
Lan ua e and Communication
Neil Postman and Charles Weingartner
Teaching
as a Subversive Activity
2.
On Propaganda, Fads, Crusades,
and Lies in General
7.
Alfred McClung Lee and Elizabeth Briant Lee,
Editors / The Fine Art of Propaganda, A
Stud of Father
8. Martin Gardner
Name of Science
9. Eric Hoffer / The True Believer: Thoughts on
the Nature of Mass Movements
10. William McGaffin and Erwin Knoll / Anything
but the Truth: The Credibility Gap -- How
the News is Managed in Washington
3. Some Particular Struggles
Between Doctrine and Reality
11. Senator J. William Fulbright / The Arrogance
of Power
12. Jerome D. Frank / Sanity and Survival
13. James E. Bristol and 7 other authors / Anatomy
of Anti-Communism
14. Mark Lane J A Citizen's Dissent
1. From M. L. Huber
E. I. DuPont de Nemours & Co.
Wilmington, De1. 19808
In your editorial "Computers, Language, and Reali ty" in "Computers and Automation", March 1970,
COMPUTERS and AUTOMATION for May, 1971
33
THE DEATH OF WALTER REUTHER
1. From Terrell L. Elrod
Systems Supervisor
Hoffman La RoChe Inc.
Nut/~y, N.J. 07110
In the past I have applauded the efforts of you
and others who have not been satisfied with the information released to the public surrounding the viol~nt deaths of our national leaders.
Supplementary
data and the analysis of events published by you
have been newsworthy and are a contribution to the
search for the true facts that surrounded their tragic deaths.
However, your January 1971 article "The Death of
Reuther: Accidental or Planned?" was put11 shed prematurely si nce the investigator, "Leonard
Walden," answered your questions in only a highly
generalized manner adding nothing to the facts already known about this incident. For example, his
comment on the altimeter was "The altimeter of the
Lear Jet could have malfunctioned eit~er accidentally or by design." How many other alternatives
are there? The important points are: did the altimeter malfunction, and if so, what was the cause;
was the altimeter tampered with, if so, by whom;
a~d.m?st of all -- if the answers to these two possibliltles are no -- was pilot error involved.
W~lter
In your apparent haste to publish an assasination hypothesis, you nowhere explore the possibility of pilot error. No mention is made concerning the possibility of unstable aerodynamic
characteristics of the Lear Jet in turbulent weather.
I strongly feel that before you can begin to
investigate a possible plot on th~ life of Walter
Reuther you must first determine that there was no
aircraft malfunction or pilot error involved.
Otherwise, we might soon start reasoning that
because more than 50,000 persons are killed on
ou: nation's highways that a possible conspiracy
eXlsts among the asphalt manufacturers of the country.
II. From the Editor
You are quite right that many aspects of the
publication of the article on Walter Reuther's tragic death would en~itle a reader to say that the
article was "published prematurely".
But there are these other factors:
(1) It
and
far
are
costs money to investigate further,
the employment of our investigator so
has cost several hundred dollars and we
not a rich magazine;
(2) We waited five months (from about July 15
to about December 15) to see if more informat~on came out by other avenues, and none
did ;.
(3) Before the sad event becomes staler still,
it was desirable to raise the issue;
(4) There exists no public agency in the United States so far as I know which is charged with investigating the deaths of our important leaders, and therefore it is incumbent on ordinary people (like "Computers
34
and Automation") to try to step into the
breach;
(5) Our act of publication might lead to more
volunteering of information.
I hope we shall have more information to publish.
A DATA BANK FOR NARCOTIC ADDICTS-- COMMENT
Bruce Madsen
10332 E. Lake Rd., RD-1
North East, Pa. 16428
Elmer Young ("Data Bank For Narcotic Addicts",
C&A, October, 1970) missed a wonderful opportuni~to use his computer tool to achieve understanding of a social problem.
Instead of using the results of studies by N.A.R.P.
(Narcotic Addict Reporting Program) to educate
computer and information engineers, he gives us statistics from unknown sources, modified by President
Nixon. He continues to confuse the discussion with
imprecise language:
" ... A significant number of marijuana users
become narcotic addicts. In a substantial number of cases ... marijuana ... leads to ...
use of harder drugs."
The issue is less clear if one compares the statements to the "data" offered:
180,000 heroine users
users -- 1 : 44.5
8,000,000 marijuana
Is one out of forty a significant number?
than 2.5% substantial?
Is less
I would like to see a follow-up article with the
goal of increasing our understanding of the narcotics
problem.
PROVOKE THINKING
Mr. Vincent K. Roach
Product Applications Manager
COMRESS
1250 Broadway
New York, N. Y. 10036
I find "Computers and Automation" an extremely
interesting publication. While some of the articles
seem to go to extremes in their viewpoint or application of "logic", this is fine in that they provoke thinking on the part of the readers. I receive
over 15 EDP-type magazines per month, and simply
can't read everything. So I mark them up for later
reading, scan to eliminate second-hand information,
poorly-documented or_-researched articles, and items
of extremely limited appeal. Those that survive,
I cut out and save. Of over 250 articles per month,
I save perhaps a half dozen. Recently, I have saved
"Computers and Automation" intact, as it contains
as much useful, interesting information as the rest
combined.
Keep up the good work. Our computer "profession"
has much maturing to do. "Computers and Automation"
is maturing with it, perhaps a bit faster.
COMPUTERS and AUTOMATION for May, 1971
Winston
Churchill
BOOKS AND OTHER PUBLICATIONS
Stuart Freudberg
Newton, Mass.
Pylyshyn, Zenon E., editor, and 35
authors. / Perspectives On the
Computer Revolution / Prentice
Hall, Inc., Englewood Cliffs, NJ
07632 / 1970, hardbound, 540 pp.,
$10.50.
would
have
loved
Xebec.
Uncommonly versatile-that
was Winnie. That's us too. When
we started in business, our aim
was to make and market versatile mini-computer peripherals
-to offer the industry quality
systems and equipment that
would be compatible with the
minis of all major OEM companies.
So we did.
Today we market such products
as advanced operating cassette
systems, moving head disk systems and controllers, line printer interfaces, and many more.
If you're looking for versatile
peripherals, contact us. Our
large staff of software and systems analysts will solve your
compatibility problems just as
we've done for hundreds of
customers throughout the U. S.
and abroad.
SYSTEMS INCORPORATED
918 North Rengstorff Avenue
Mountain View, Calif. 94040
Telephone 415-964-4263
TWX 910-379-6942
COMPUTERS and AUTOMATION for May, 1971
Contains a collection of important readings on many aspects of
"the computer revolution", and includes statements on the historical,
practical, theoretical, philosophical, social, legal, and moral issues
of the computer field, from many
distinguished authors.
The book is divided into three
sections: (1) "The Development of
Computer Science," which describes
the development of computers and
the intellectual heritage of computer science; (2) "Man and Machine"
which explains the relationship between man, as a conscious thinking
organism, and the machine; (3) "Society and Machine," which examines
the relationship between society as
a whole and the machine, which includes the impact of computers and
related technological phenomena on
our society.
Taviss, Irene, editor, and 30 authors / The Computer Impact /
Prentice-Hall, Inc., Englewood
Cliffs, NJ 07632 / 1970, hardbound,
297 pp., $7.95.
Using writings of many authors,
this work examines the social implications of computers by looking
at their potential and at their impact on the economy, political organization, and the culture of society.
Sackman, Harold J Man-Computer Problem Solving / Auerbach Publishers,
Inc., 1101 State Rd., Princeton,
NJ 08540 / 1970, hardbound, 272
pp., $12.50
This book is an examination of
the growing experimental evidence on
man-computer problem solving, particularly the competition between
time-sharing and batch-processing
computer systems. The author examines how-people solve problems
and how they can use computers most
effectively to reach their solutions.
Gallagher, Cornelius E., New Jersey.
Chairman, House Subcommittee on
Invasion of Privacy, and others /
The Computer and Invasion of Privacy: Hearings Before a Subcommittee of the Committee on Government Operations. House of Representatives, Eighty-Ninth Congresf
July 26 to 28, 1966: Reprint /
(Please turn to page53/
NEW DIRECTIONS
IN COMPUTER
PROGRAMMING FROM
WILEY-INTERSCIENCE
PL/I PROGRAMMING
IN TECHNOLOGICAL
APPLICATIONS
By Gabriel F. Groner, THE RAND
CORPORATION
"This book was written for those who
want to quickly learn to solve engineering and scientific problems by writing
computer programs in Pl/I (Programming
language One). It gets the beginner off
to an easy start, but treats some advanced
topics as well. It may be used profitably
as a text in a computer science course for
engineering, science, and mathematics students, as a supplemental text in a course
emphasizing problem solving, in an industrial course, or as a self-teaching guide in
individual study." - from the Preface
1971
240 pages
$9.95
SYSTEM/360 JOB
CONTROL LANGUAGE
By Gary DeWard Brown, THE
RAND CORPORATION
This manual presumes no knowledge of
System/360 JCL and is appropriate for
those familiar with any computer language
whether they code in COBOL, FORTRAN,
PL/I, assembly language, RPG, or some
other language. The manual serves as a
learning text for the programmer who
wants to understand and use System/360
Job Control language, and as a reference
for the experienced JCL programmer.
Each Job Control language feature is
described in complete detail, examples
are given for its use, and possible applications are discussed.
1970
292 pages
paper, $7.95
AUTOMATIC DATA
PROCESSING
System/360 Edition
By Frederick P. Brooks, Jr., UNIVERSITY OF NORTH CAROLINA
AT CHAPEL HILL, and Kenneth E.
Iverson, THOMAS J. WATSON
RESEARCH CENTER, IBM
Of direct interest to data processing
specialists and to workers in all fields,
this outstanding volume covers the fundamental aspects of data processing common
to all fields of application. It illustrates
and applies theoretical material solely in
terms of IBM's System/360 computers.
Since Professor Brooks managed the design of the System/360, this is an especially authoritative introduction to machine
principles and functions.
1969
466 pages
$14.50
WI LEY-INTERSCI ENCE
a division of JOHN WILEY & SONS, Inc.
605 Third Avenue,
New York, N.Y. 10016
In Canada:
22 Worcester Road, Rexdale, Ontario
35
LYING BY THE UNITED STATES GOVERNMENT:
AN "ACCEPTABLE LEVEL"?
Edmund C. Berkeley
Editor, Computers and Automation
Every now 'and then these days I find out something that lacerates and splinters all my hopes
and desires to have a good opinion of. an admiration for. the present United States Government.
and its "national security" arm. the Pentagon and
the military. In the years 1942-1946. I was on
active duty in the U.S. Naval Reserve. and my last
rank was Lieutenant Commander; and in those years,
how much I admired the U.S. Navyl
I have wondered for a long time how reliable
are the figures issued by the computers in the
Pentagon about the number of deaths of Americans
and Asians as a result of the war in Vietnam. The
battle death toll of Americans is currently reported by the Pentagon to be somewhere between 50,000
and 55,000.
I happen to believe that the citizens of the
United States, a democracy. are entitled to "the
truth. the whole truth. and nothing but the truth"
from the government which they have chosen to serve
them.
Apparently, the truthful battle death toll (including the toll of the aftermath of battle and
of the destruction of a soldier into a living
corpse) is at least 200,000 Americans.
Fol~owing is an excerpt from a most important
book, "Conversations with Americans" by Mark Lane.
pub~ished by Simon and Schuster (630 Fifth Ave.,
New York, N.Y., 1970. 247 pp). Peter Sc~wed, the
president of Simon and Schuster, told me in converiation that they had most carefully verified the
information in the book before they published it.
Most of the book consists of verbatim testimony
from 32 Vietnam veterans, accounts based on tape
recorded interviews of how the U.S. Marine Corps
and other military groups trained them in torture
tactics, and of the atrocities and massacres these
soldiers witnessed or participated in. But in the
introduction there occurs the following report:
"If Americans know less than all there is to know
about the terrible cost the war is imposing upon the
civilian population of South Vietnam, they know next
to nothing of the real cost America is paying for its
adventure.
"The real price is in the sacrifice of an entire
generation.
"Some of the untold stories may be gathered in the
hospitals in Japan, Germany, and the United States.
Major Thomas Engelsing, now chief of the in-patient
service in psychiatry at the Army's 97th General
Hospital in Frankfurt, told me that as a result of
immediate medical contact with the wounded in Vietnam
Lives are saved in this war. Lives of men who
in any other war would have died. Men with
multiple amputations, blinded,very serious
brain inj uries.
As a result,
the Army facilities are now overfilled with
men who are not rehabilitable.
36
He said that he had received
a directive not to evacuate wounded to the
United States because they are just filled
up in the States; there is no more room.
"Major Engelsing said that the facilities in
Germany are so overcrowded that he has had to
ignore the directive. A colonel who had just
returned to Germany from the United States had told
him that conditions in Army hospitals and Veterans
Administration hospitals in the U.S. are so deplorable and the facilities so overcrowded that a substantial number of doctors and nurses, unable to
stand the depressing scene any longer, have asked
to be transferred to Germany or even to Vietnam.
Conditions in Army and Navy hospitals in Japan
are similar --perhaps slightly worse.
"The Administration nevertheless has found a new
method for reducing casualty figures. A man with
both arms and legs blown off and suffering incurable brain damage is listed by the Pentagon merely
as "wounded." When he dies in an Army hospi tal in
the United States or in Japan or Okinawa he becomes
a domestic military death -- not charged to the war,
not added to the total. In this fashion U.S. losses
in Vietnam are kept at an acceptable level, to
borrow a Pentagon term. But for each man who dies
on the battlefield in this war, doctors and medics
have told me, another probably dies elsewhere. And
for each who has died, perhaps one or two others
will be unable to functiori ever again. The official
figure for American deaths in Vi~tnam is over
45,000. But a figure several times that would more
accurately reflect the number of young men lost
forever. Yet even that figure would fail to comprehend the permanent damage done to so many who
have fought in Vietnam who have not been physically
wounded."
(End of excerpt.)
Some confirmation in general of what Mark Lane
here says may be found in an article in "The Washington Monthly" for April 1971, (published at 1150
Connecticut Ave., N.W., Washington, D.C.) This
article is entitled "The Burn War" by Ronald J.
Glasser, a major, 1968-1970, in the United States
Army Medical Corps. Glasser more or less matter
of factly tells the story of a burned and dying
American soldier, David Grant, in an American-run
hospital at Kishine, Japan. Grant, 20 years old,
was carrying detonators in his rucksack, and they
blew up, and put him on fire. The half dozen pages
describe the death of David Grant from burns and
bacterial infection. After his savage suffering
and death, this boy is NOT counted as a battle
death.
Every information system, computerized or not
computerized, is foully contaminated by lies.
To say that American deaths from fighting in a
war are 50,000 when they are actually over 200,000
is a lie.
Americans must not continue to tolerate lying
by their government. The persons who lie, and the
persons who cooperate in the concealment and the
lying such as President Richard M. Nixon, who is
responsible for knowing the "over 200,000" figure,
must be voted out of office, or otherwise dismissed from the government of the United States.
COMPUTERS and AUTOMATION for May, 1971
LI FE SUPPORT AND WAR CONTROL SYSTEMS
FOR PLANET EARTH
Howard Kurtz
War Control Planners
Box 35
Chappaqua, N. Y. 10514
Without danger to the national security of the
United States, the President of the United States,
with strong bi-partisan support in Congress, can
launch new strategic initiatives to chart a new historic course for world civilization:
All nations can be invited to direct a portion
of their research and development budgets toward a cooperative war prevention decade.
No risk to the security of any nation will be
incurred, because no nation will be asked to turn
over its defense to the experimental global safety
organizations.
The experiments can go forward with U.S. initiatives, no matter which nations join or hold back at
present. Matters can be so arranged that any nation
can enter into the experiment at any later date.
This mobilization of American creativity for world
leadership will be in addition to continued essential
defense efforts. The following tentative description
of a large-scale strategic initiative is offered: to
be debated; to be improved; to stimulate the creative
thought of others, ~ithin their special fields.
1. Global Safety and Development Services
Military and technical and intelligence experts
of all cooperating nations will work together developing and testing military surveillance capabilities.
All information will be unclassified. Knowledge and
experience will be accumulated pertinent to that
future time when a world-wide peace-keeping authority
will be in existence. It will be assumed that individuals will distort information due to political
or national loyalty, and this will be a decade of
experiment in multi-national checks and balances to
correct for such distortions. It also will be a
decade of experiment with increasing capacities and
inadequacies of computers and cybernetic systems, in
the evaluation of global safety intelligence.
2. Secrecy
Any nation may try to keep secret any information
it considers essential to its security, but it will
not be able to prevent other nations from making public such information if they are cognizant of it.
The Soviet Union, for example, could make public to
the world any information it has about American military activity, or the United States could make public similar information about the military activities
of all other nations, from intelligence gathered
through any source.
Although this ten-year prototype experiment will
be inaugurated with information garnered from Earthorbiting satellites and high flying airplanes, nations
will be able to provide input from all intelligence
channels.
3. Incidents or Crises
The United States will announce the commitment of
a special ten-year continuing series of ne~ earthorbiting Satellites and associated systems (computerized and other) to the future safety and wellbeing
of mankind.
Whenever border incidents or war between nations
occur. close-up surveillance of both sides, day and
night. can supply the world's public with current
facts of alleged or actual hostile actions. Audiovisual information from the centers may be transmitted onto home television screens in times of special
emergency, by the flip of a switch.
A giant open-to-the-public information display and
model global safety and development management center
~ill be built somewhere near the United Nations Headquarters, ~here the intelligence information garnered
by these systems ~ill be fully available to the press
and the public.
All nations having the capability to launch Earthorbiting satellites will be invited to provide their
share of scheduled launchings. Otherwise the experiments will go onward with U.S.-launched satellites.
All cooperating nations can build identical opento-the-public management centers.
These ~ill be in continuous communication and will
operate simultaneously through a global information
exchange network.
These satellites, associated systems and ground
support centers and computers will have t~o principal
functions:
First, to maintain public inventory of all
potentially dangerous military activity for
planet Earth. This will be a new, additional
level of world military intelligence, separate
from each nation's military intelligence, and
serving as an additional potential safeguard
against a war starting by mistake, or misunderstanding.
Second, to assist the economic development and
human wellbeing of all nations by providing
information of all types to nations on a continuing and immediate basis.
COMPUTERS and AUTOMATION for May, 1971
All nations will be invited to design and produce
intelligence-gathering instruments to be placed on
these experimental "eyes in the sky." A new kind of
"race" will be launched in which all nations. large
and small. can make contributions to the future
safety and wellbeing of mankind, in contrast to the
race to produce powers of violence to destroy world
ci viliza ti on.
4. Non-Military Benefits
Perhaps the greatest benefits will come from the
non-military uses of global public inventory services.
In the United States. the National Aeronautics and
Space Administration, the Department of Agriculture.
the Department of the Interior. the Geologic Survey.
the Department of Fisheries. the Federal Aviation
Agency and many other departments and agencies already have experiments in progress for a series of
Earth Resources Technology Satellites (ERTS) to be
launched beginning in 1972 in Earth orbit.
Through the global information centers and other
channels these powers being created in the 1970s will
?e utilized to serve mankind, rather than in the sole
Interests of the United States. Experiments are in
progress to use satellites for air and sea and land
37
navigation and traffic control, and for search & rescue and emergency help anywhere in the ~orld. In
time, even the smallest lifeboat in the largest ocean
will be located by these new systems, to save human
Ii ves.
Experiments now under way will make it possible
to maintain inventory of overy kind of agricultural
crop and herd, in every country in the world, every
day, from outer space. Other experiments in process
will make it possible to determine the chemical content of surrounding soil, and its moisture content,
and therefore predict the quality and quantity of all
crops for the future global markets. Global meteorological services already are·in existence using
photography from outer space.
Satellites will be used to locate the largest
schools of fish in the oceans of the world and to
direct fishing fleets to them by radio, to facilitate feeding the hungry of the world.
Information will be made available to all nations
to monitor the pollution of the air and water
throughout the planet, daily. These same sensors
will detect illicit military production as 'it
changes the pollution in its environment. Geologic
surveys from outer space will help nations locate
oil and mineral and other natural resources. Global
forestry management services and water management
services will help all nations toward maximum productivity to meet their human needs.
5. The Changed Character of
Government-Sponsored Research
Faculty, student and public support for government-sponsored research will grow to encourage the
pioneering of peace technology •.. war safety control
intelligence ••• war prevention control experiments •..
global development services •.. with research and
development directed toward pro-human purposes.
All cooperating nations will be invited to participate in continuing series of war prevention
games, seminars, workshops and conferences. (Where
war games test the proficiency of nations in destroying each other, war prevention games will test
the proficiency of nations in protecting each other
from war.)
6. War Prevention Colleges
New kinds of multi-national military colleges
and war prevention colleges will study future law
enforcement structures to enforce world safety law,
and to maintain lasting world peace, instead of to
wage world war.
New kinds of multi-national engineering institutes and strategic "think factories" will pioneer
the technical aspects of planetary armed forces
management systems, eventually to bring an end to
war.
New kinds of economics schools and business colleges will project new concepts of global organization to manage Earth-orbiting satellites and other
global services, in the interest of all nations,
rather than in the interest of the superpowers.
New kinds of political science colleges and
schools of diplomacy will study unprecedented
checks and balances and controls for the new world
organization, or reconstituted Uniied Nations of the
future, in which all nations will find national
security ••. and in which all nations will find national independence.
7. A New American Purpose
After World War I the League of Nations was
established to bring an end to war. It was not
given authority or power to achieve this objective.
Then after World War II the United Nations was
established to bring an end to war. It was not
given the authority or power to achieve this objective. If we wait until after World War III to
create the global public authority capable, in fact,
of bringing an end to war ••• it will be too late.
One thing is certain: no progress will be made
through present global strategies of the two superpowers or until a massive research and development
and testing commitment is made, and maintained,
firmly committed to this goal of a future free from
war.
No risk to the national security is involved in
the experiments which art herewith proposed.
THE GROWTH OF THE COMPUTER INDUSTRY
James B. Hunter, President
Digital Scientific Corp.
San Diego, Calif.
We predict that the computer industry will double in size every three years in the near-at-hand
future, but the number of companies in it will be
reduced by half.
Our projections indicate that our industry will
grow at a rate of at least 30% per year, or slightly
more, for at least the next several years. Thus it
will be twice today's size in just about three years,
and we expect it to continue to grow at that rate."
We also predict a drastic shakeout, expecially
among manufacturers of so-called "mini-computers."
Today there are around 100 companies making small
computers. We expect there will be room for 20 or
30 at the most.
We expect the greatest portion of qrowth in the
computer industry to be in "midi-computers." A
"midi-computer" is a high-speed computer which will
operate alone or in concert with very large machines.
"Midi-computers" apply to communications processing, controlling high-performance peripheral
devices, and many other tasks for science and business.
Based on our record to date, we expect to grow
faster than a 3Q%-a-year rate. Our company is just
over three years old; it has jumped from an annual
sales rate of $4,000,000 last year to an anticipated
$10 million in 1971. We see no reason why we can't
continue or even increase this rate of growth.
New kinds of law schools and bar associations
and world law institutes will continue to pioneer
new concepts of inter-nation conflict law, and
undersea resource management law, and other unprecedented imperatives for planet-wide law, and
enforcement power.
38
COMPUTERS and AUTOMATION for May, 1971
CALENDAR OF
COMING EVENTS
May 3-5, 1971: Data Processing Supplies Association, Affiliate Membership Meeting, Copenhagen, Denmark / contact: Data Processing
Supplies Association, 1116 Summer St., Stamford, Conn. 06905
May 5-6, 1971: 16th Annual Data Processing Conference of Univ. of
Alabama Graduate School of Business and Division of Continuing
Education, Parliament House Motor Hotel, 420 South 20th St., Birmingham, Ala. / contact: C. E. Adams, Director of Conference Activities, Box 2987, University, Ala. 35486
May 11-13, 1971: IEEE (Institute of Electrical and Electronic Engineers) 1971 Region Six Conference, Wood lake Inn, Sacramento,
Calif. / contact: Dr. D. H. Gillot, Co-Chmn, IEEE Region 6 Conference, Sacramento State College, Dept. Of Electrical Engineering,
6000 Jay St., Sacramento, Calif. 95819; or, Dr. R. F. Soohoo, Program Chmn., IEEE Region 6 Conference, Univ. of California at
Davis, Dept. of Electrical Engineering, Davis, Calif. 95616
July 26·29, 1971: First International Computer Exposition for Latin
America, sponsored by the Computer Society of Mexico, Camino
Real Hotel, Mexico City, Mexico / contact: Bernard lane, Computer
Exposition, Inc., 254 West 31st St., New York, N.Y. 10001
Aug. 3·5, 1971: ACM '71 "Decade .of Dialogue", Conrad Hilton Hotel,
Chicago, III. / contact: AI Hawkes, Computer Horizons, 53 West
Jackson Blvd., Chicago, III. 60604
Aug. 3-6, 1971: IFAC Symposium on The Operator, Engineer and Ma~.
agement Interface with the Process Control Computer, Purdue University, lafayette, Ind. / contact: Dr. Theodore J. Williams, Purdue
laboratory for Applied Industrial Control, Purdue University, lafay·
ette, Ind. 47907
Aug. 11.13, 1971: Joint Automatic Control Conference, Washington
Univ., St. louis, Mo. / contact: R. W. Brockett, Pierce Hall, Harvard
Univ., Cambricge, Mass. 02138
May 12-14, 1971: Annual Regulatory Information Systems Conference
of the Missouri Public Service Commission, Chase-Park Plaza Hotel,
St. louis, Mo. I contact: Sam l. Manley, Secretary, Missouri Public
Service Commission, Jefferson City, Mo. 65101
May 12-14, 1971: 22nd Annual Conference of the American Institute
of Industrial Engineers (AilE), Boston, Mass. / contact: Anthony J.
Jannetti, Exhibit Manager, c/o Charles B. Slack, Inc., Pitman, N.J.
08071
May 18-20, 1971: Spring Joint Computer Conference, Convention Ctr.,
Atlantic City, N.J. / contact: AFIPS Headquarters, 210 Summit Ave.,
Montvale, N.J. 07645
May 21-22, 1971: Communications Systems Management Association
First National Meeting, New York City, N.Y. / contact: CSMA Convention, P.O. Box 2805, Wilmington, Del. 19805
May 24-26, 1971: Power Industry Computer Applications Technical
Conference, Statler Hilton Hotel, Boston, Mass. / contact: P. l.
Dandeno, Hydro Electric Power Commission of Ontario, 620 University Ave., Toronto, Ontario, Canada
May 24-28, 1971: 2nd International IFAC Conference and Exhibition
"P.R.P.-Automation", Centenary Halls, Brussels, Belgium / contact:
IFAC/P.R.P.-Automation, Jan van Rijswijcklaan 58, B-2000 Antwerp,
Belgium
June 1.4, 1971: Seventh Annual Data Processing and Automation Con·
ference, National Rural Electric Cooperative Association, The Riviera
Hotel, Atlanta, Ga. / contact: C. E. Aultz, NRECA, 2000 Florida Ave.,
N.W., Washington, D.C. 20009
June 2·5, 1971: 3rd IFAC/IFIP Conference on Digital Computer Applications to Process Control, Technical University, Otaniemi, finland / contact: 3rd IFAC/IFIP Conference, Box 10192, Helsinki 10,
Finland
June 3·5, 1971: Conference on Area-Wide Health Data Network,
School of Medicine, State Univ. of New York at Buffalo, Buffalo,
N.Y. / contact: Continuing Medical Education, 2211 Main St., Buffalo, N.Y. 14214
June 7-9, 1971: International Computer Forum and Exposition (ComFor), McCormick Place-on-the-lake, Chicago, III. / contact: National
Electronics Conference, Inc., Oakbrook Executive Place ", 1211 W.
22nd St., Oak Brook, III. 60521
June 21-22, 1971: Ninth Annual Conference of the Special Interest
Group on Computer Personnel Research of the Association for
Computing Machinery, Center for Continuing Education, Univ. of
Chicago, III. / contact: Fred A. Gluckson, EDP Systems Dept.,
National Bank of Detroit, Detroit, Mich. 48232
PROBLEM CORNER
Walter Penney, CDP
Problem Editor
Computers and Automation
PROBLEM 715: A RUN AROUND THE BASES
When Claude Liffey entered the classroom, he saw John
Lawthorne at his desk studying a pageful of calculations.
"That's not right, is it," he said, pointing to an example
John had just finished writing: 10.01 x 100.01
1010.0001. "Shouldn't the answer be 1001.1001 since
2 1/4 x 4 1/4 is 9 9/16?"
"That would be right if this were binary, but it isn't.
That is, 1 and a are the only digits used, but the base isn't
2."
"Then it probably isn't positional. Something like
Roman Numbers?"
"Not at all, it's completely positional, on both sides of
the point", John said. He continued in the manner of a
sightseeing guide. "On the left we have powers of the base
and on the right we have reciprocals of the base and its
powers."
"What is all this then?"
"Well, it's an exercise I was thinking of using in my class.
We're studying number systems now and I thought I'd give
them a change from the usual bases and develop some really
far-out system for them. Look at this example," he said,
poin ting to 1000.100 1 + 1010.000 1 = 1010 1.0 1al. "Would
you believe these are all integers?"
"Too much for me. Do you expect your students to
figure this out?" asked Claude.
"Well, yes and no. I don't think anyone could determine
the base used." Is this true?
Solution to Problem 714: The Numbers Game
July 19.21, 1971: 1971 Summer Computer Simulation Conference.
Sheraton-Boston Hotel, Boston, Mass. / contact Dona Id H. Niesse,
McDonnell Automation Co., Dept. K676, Box 516, St. louis, Mo.
63166, or, Peter Stein, McGraw-Hili Publishing Co., 607 Boylston
St., Boston, Mass. 02116
He started with 3249, getting 0576 as the result of the
first operation. The numbers 24 and 57 then repeated
endlessly.
July 19·23, 1971: Conference, on Computers in Chemical Education and
Research, Northern Illinois Univ., DeKalb, III. / contact: Dr. F. M.
Miller, Dept. of Chemistry, Northern Illinois Univ., DeKalb, III. 60115
Readers are invited to submit problems (and their solutions) for
publication in this column to: Problem Editor, Computers and
Automation, 815 Washington St., Newtonville, Mass. 02160.
COMPUTERS and AUTOMATION for May, 1971
39
ACROSS THE EDITOR'S DESK
35-LB. COMMUTER COMPUTER
UPS TRAFFIC FLOW 500%
IN NEW YORK CAPITAL
was put into effect on March 15.
On that date, notification and official input forms were mailed to
over 25,000 priests and pastors of
three leading denominations and the
National Council of Churches.
A minicomputer-controlled network of traffic signal s has increased the flow of cars and trucks
in the business di strict of Albany,
New York, from 1,000 cars per hour
to nearly 5,000, while saving a
significant amount of municipal
money in manpower and congestion
costs. The system, called RAPIDTRAC, uses Varian Data Machines'
620/i digital minicomputer.
The mi ni ster' s input form now
being disfributed asks for a wide
range of information, incl uding interests and skills. This information will be fed into the computers
of the various denominations and
the National Council of Churches.
Each participant then will receive
a printout so that he will be able
to review and approve hi s own record.
Designed specifically for Albany's Central Business Oi strict by
Computran Systems Corp., Hacken sack,
N.J., RAPIDTRAC -- for Real-Time
Automatic Programmed Intersection
Digi tal Traffic Cont rol -- has provision for such interrupts as 'fire
preempt'. Fire stations in the vicinity can use small consoles within the fi rehouse i tsel f to interrupt the normal flow of traffic
and command the 620/i to optimize
the fire equipment's travel through
downtown intersections.
Beginni ng next fall, any congregation of the Episcopal, American
Baptist or Lutheran Church in America that has a vacant pastorate may
apply to its national headquarters
data bank and receive personnel
profiles on ministers looking specifically for its kind of parish.
The congregation will then follow
up by conducting interviews and
making its own selection.
APPLICATIONS
The minicomputer accepts signal s
from a network of 65 traffic detection loops buried in the road
pavement at selected locations.
The 620/i' s memory identifies and
records measurements of the tr"lffic
on a continual "trend" basi s, and
al ters the timing of each traffic
signal to optimize the flow of
traffic through the two-square-mile
network. Typical cycle time of a
signal is &) seconds, maintaining
about a 30-m.p.h. synchronized flow
of traffic.
The system, designed with the
prime consideration of safety, can
adjust the delicate rhythms of
traffic signals according to wet,
icy or snowy weather condi tion s, as
well as for unusual fluctuations in
traffic density such as holidays,
parades and various civic observations. The Varian 620/i provides
RAPIDTRAC wi th automatic status reports on command. It al so provides
an interactive information retrieval capabil i ty vi a a tel etype terminal for traffic engineers' communications.
The system, designed by Information Science, Inc., New City,
N. Y., is programmed primarily in
COBOL to operate on IBM 360/30 (65
K) under DOS.
Each denomination
will maintain its own separate,
but compatible, computer system.
While they will use the same computer programs, each will maintain
its own data base and have complete
control of its own information.
In addi tion to pari sh mini sters,
the system will contain data about
other personnel, such as agricultural, medical and other missionaries. It also will be used to assist in educational program planning, salary research studies and
career counseling.
OTC QUOTE SYSTEM
ACTIVATED BY BUNKER-RAMO
CHURCHES ANNOUNCE NEW
PERSONNEL INFORMATION SYSTEM
NASDAQ, a new system that makes
OTC
s to c k quo tat i 0 II sill S t ,. n tl y
available on a nationwide basis, became full y operational in February,
reporting bid and ask prices on
2, :~oo unli sted issues. The quotes
of numerous market makers deal ing
in each stock are available in 700
trading rooms coast-to-coast through
the new compu teri zed network developed by Bunker-Ramo Corporation,
Stamford, Conn.
How can a congregation and the
right pastor or priest find each
other? The answer is the computerized Church Manpower System whiCth
The system employs dual UNIVAC
1180's, 8 modified Honeywell 5l6's,
30,000 miles of high-speed lines,
and 1,500 CRT terminals. The cen-
40
tral computers calculate median
quotes on all issues and these are
available to investors through thousands of brokerage offices which
already have subscribed to the new
service through their existing quotation systems.
Special Bunker-Ramo cathode ray
tube terminal s are used to look up,
enter and update their bid/ask
quotes. Unlike an exchange-traded
stock which is quoted by a single
specialist, an OTC stock is quoted
by up to 40 market makers, each of
whom may have a different bid and
offer. NASDAQ enables trading room
personnel to see all such quotes on
any issue in a few seconds by entering a stock symbol on the terminal keyboard. Brokerage office
personnel, through Bunker-Ramo's
Telequote III or other similar services, obtain median NASDAQ quotes
by a simple keyset interrogation
simil ar to that performed for li sted
stock quotations.
The $25,000,000 system is operated by Bunker-Ramo under a contract with the National Association
of Securities Dealers, a regulatory
body for all over-the-counter securities trading. NASDAQ is an acronym for National Association of
Securi ties Dealers Automated Quotations, and is a trademark of the
NASD. NASDAQ is expandable to accommodate up to 10 ,000 stock sand
to perform other functions for the
sec uri ties industry such as the recording and clearing of all trades.
TRANS WORLD AIRLINES
IS TESTING COMPUTERS
WITH COMPUTERS
Trans World Airlines announced
recently that one of two Bendix
automatically controlled computerized test stations has been put
into operation at TWA's Technical
Services Center, Kansas Ci ty International Airport. The install ation
is the fi rst maj or step taken to
apply computer science to the testing of the advanced state electronic equipment on the airline's
growing fleet of Boeing 747 widebodied jet s.
The new equipment is currently
programmed to automatically test
generator panels, flight steering
computers and the PB20 automatic
pilot, all found aboard the Boeing
707's or 727's. According to J.T.
Davi s, staff vice president of maintenance and overhaul. In addition
to a sharp reduction in test time,
adVantages of the new test station,
incl ude a check-out of equipment to
COMPUTERS and AUTOMATION for May, 1971
a far greater extent and wi th greater
assurance of complete reliability.
An auto-pilot amplifier, for instance, is being checked out in less
than two hours using the new equipment. Formerly, testing time could
run as much as a full working day
on each auto-pilot.
When fully implemented into TWA's
maintenance operation, the Bendix
test stations will be capable of
checking out all auto-pilot and
flight director type equipment and
will extend the exi sting capabili ty
for testing digital avionics. The
test stations, Bendix series model
260 and 270, also will be used for
testing of communications and radio
navigational aid avionics.
TIGHTER CONTROL OVER GUNS
PROVIDED BY IBM COMPUTER
FOR NATIONAL DISTRIBUTOR
Dayton Gun Headquarters (Ohio),
a national distributor of firearms,
hunting supplies, fishing tackle
and archery equipment,has insti tuted
further safeguards for the control
and di stribution of fi rearms with
the installation of an IBM System/3
Model 10. The Model 10 provi des
more accurate record keeping and
faster access to gun serial numbers.
Under the previous manual system, the chance of transcription
errors taking place as serial numbers were recorded was always present, and the manual process was
time consuming. With the new system, as guns come in from the manufacturer, the comprehensive inventory (including a listing of all
seri al numbers) is entered into the
computer. When i terns are sold, the
computer automatically deducts the
indi vidual uni t s from inventory.
Records of the guns shipped to a
particul ar deal er al so are recorded
in the computer. Before the shipments are made, a final visual check
of serial numbers is made to assure
that all units are accounted for.
In the event a seri al number inqui ry
is received from some law enforcement agency, the computer can quickly
identify the appropriate dealer
that received the units.
The IBM system al so reviews each
order pI aced wi th the company and
checks it against a master list of
1,500 gun dealers to insure the
customer receiving the order is licensed for gun sales. In the event
a dealer's license either is not
registered or has expired, the computer will omit the guns from the
order and print a notation on the
invoice telling him why guns are
not being shipped.
"In many cases," Lee Brock,
treasurer of Dayton Gun, says, "a
COMPUTERS and AUTOMATION for May, 1971
dealer simply forgets to send us a
copy of his renewed license. But
this system puts the burden of proof
on the government to issue the license and the dealer to let us know
of his certification." The system
also provides a number of financial
support and inventory control functions for the company.
DETROIT AREA RETAILERS CAN
NOW SELECT STORE LOCATIONS
WITH AID OF A COMPUTER
Retailers in the Detroit area,
planning to establish new outlets,
now can receive computer assistance
to help them pick the best location
to attract maximum customer trade.
The servi ce, operated by the Detroit News (Michigan), is made possible by a large-scale Sperry Rand
UNIVAC 1108 computer system in
Chicago and Computer Sciences Corporation's nationwide time-sharing
network known as INFONET. Current
and potential advertisers in the
Detroi t News can use the service,
which is known as a Pilot Site Analysis Program.
The program estimates the number
of households, total gross consumer
income and dollar value of the total
commodi ty market wi thin a fi ve mile
radius of a proposed retail site.
This trading area is subdivided into
nine smaller segments. Households,
income and commodity estimates and
proj ections for each of the nine
segments are based on dwelling unit
intensity with each area. Results
are then added to provide totals
for a trading area.
To perform the analysi s, the computer employs some programs already
in its "library" plus others especially prepared by the Detroit News
staff. Codes representing data on
each of the nine trading area segments are transmi tted over leased
telephone lines from the Detroit
·News terminal to the Uni vac computer.
Wi thin 20 minutes, the total data
compri sing a pil ot si te anal ysi sis
printed out on a low-speed printer.
The analysi s i s photocopied with
accompanying maps and bound into a
report. Normally such an analysis
would cost a retailer more than
$1,000 if he used a consultant. It
costs the Detroi t News $7 and the
retailer gets it free as a news
service. To date, more than 76 pilot site analyses have been run by
the newspaper.
EDUCATION NEWS
TWO RCA COMPUTERS HELP TEACH
APPALACHIAN-AREA ELEMENTARY
AND HIGH SCHOOL STUDENTS
The Eastern Kentucky Educational
Development Corp. recently dedicated
a new RCA computer installation to
help teach almost 2,000 Appalachianarea el ementary and high school
students. A total of 34 teletypewri ter computer terminal s are located in 26 semi-rural and rural
publ ic and parochial school s, the
Kentucky School for the Deaf and
the Federal Youth Center, a correctional institution for boys.
All 34 termi nal s are 1 inked by tel ephone lines to an RCA Spectra 70/45
and an RCA 1600 computer, both of
which are housed at the Holy Family
Catholic School, Ashland, Kentucky.
According to EKEDC Director Edwin R. Jones, the CAl program is
similar to those now operating in
such metropolitan areas as Cincinnati and New York City. He noted,
however, that this is probably the
first time it has been attempted in
an area where participating school s
are so widely scattered.
The EKEDC is a non-profi t cooperative serving 32 school districts. Most of the school s are
anywhere from 50 to 200 miles a~ay
from the EKEDC computer center,
wi th some of them being in such remote areas that telephone 1 ines had
to be installed before the terminal s
coul d be connected. Each tel etypewri ter terminal can accommodate approximately 60 students per day and
offers an individualized "give and
take" relationship wi th every child.
"By individualizing instruction according to a child's level of ability, it provides the motivation
needed by so many of our socially
disadvantaged to compete in a world
for which they are otherwi se so
poorly prepared," Mr. Jones said.
Besides computer assisted instruction, the RCA computer install ation al so will be used to
process such school district admini strative data as student schedules,
grade report s and teacher payroll s.
KIDS AND COMPUTERS
At Polytechnic Insti tute of Brooklyn (New York) 14 typical youngste:r;s -- not geniuses -- are receiving a thoroughly modern education,
and having fun. The students
fifth, sixth and seventh graders at
Anna C. Scott Elementary School in
Leonia, N.J. -- travel to Polytechnic's downtown Brooklyn campus
every Saturday for their two-hour,
non-credit weekly course.
41
" ••• One of the main reasons for
our program is to insure that these
kids "don't develop the feeling of
awe toward computers that is imbedded in most adults," said Dr. Norbert Hauser, head of Polytechnic's
Operations Research and System Analysi s Department. "We want them
to get a basic understanding of
what a computer can do -- to learn
something about computer technology,
modeling, mathematics and programming." Dr. Hauser, who donates hi s
time, and two paid assi stant s are
conducting the 10-week program.
The course was arranged through
the elementary school's Home and
School Association.
Norman Falk,
a Home and School Association member who drives the children to Polytechnic every week, said the members feel that thi s supplement to
a normal elementary school education will have far-reaching and
beneficial results -- not only for
the 14 students enrolled, but for
the entire educational process at
the elementary school level.
Having learned an elementary
computer 1 anguage known as 'BASIC' ,
the students use it to work math
problems -- such as finding square
roots -- and inventing random number guessing games in which the
computer is programmed by them to
answer correctly.
Probably the
most ambi tious proj ect undertaken
by any of the students has been an
attempt to write a program for estimating the number of cars which
cross the George Washington Bridge
in one day.
Mr. Falk said, "This boy had to
take into consideration such factors
as how many lanes are in use, how
fast the vehicles travel, and how
long the cars wait at toll booths.
The problem finally became so complicated that he couldn't finish it.
Right
now,
he's programming an
imaginary game of Russian Roulette."
MOBILE OIL DONATES
IBM SYSTEM TO PURDUE
Donation of an IBM 7094-1401
computer system by the Mobil Oil
Corporation to Purdue University's
Computing Center recently was announced jointly by the corporation
and the uni versi ty. Prof. SaUl Rosen, director of the Purdue Computing Center, said the contribution
provides "an added link in a unique
mul ticomput"er system." The Computing Center already includes two
other 7094' s,
two 1401' s and a
third-generation CDC 6500, the key
instrument in the Purdue system.
Each of the 7094's has a memory of
32,768 36-bit words, or data units,
while the 6500 has a memory of
98,304 6O-bit words. Purdue now
42
uses a 7094 to drive more than 60
terminal s in a system called PROCSY
(Purdue Remote On-line Console System).
Rosen said that one of the direct benefits to Purdue from Mobil
Oil's gift would be the expansion
of PROCSY to more than 100 terminal s. The expansion will be achieved
by using two 7094' s as front end
machines, while continuing to run
Purdue's PUFFT system on a third
7094. PUFFT isafast FORTRAN compil ing system developed at Purdue
to run student jobs on the 7094.
I
At thi s time more than 2,000
such jobs are run every day and a
7094 could easily run twice that
number, Rosen noted. More than 100
courses on Purdue's main and regional campuses, including courses
in such diverse areas as engineering, English, psychology and even
home economics, regularly have assignments that require computer use.
RESEARCH FRONTIER
MAP MAKING TIME REDUCED BY
DIGITAL CARTOGRAPHIC SYSTEM
An experimental system for dramatically reducinh the time involved
in color map making was described
recently by an IBM engineer at the
American Society of Photogrammetry
Convention. The system uses a drumtype optical scanner/plotter, an
IBM 1130 computer, and an IBM 2250
Model IV display unit to produce
separation negatives for multicolor
maps. It can reduce engravi ng time
for map generation from months to
as little as one day.
The system first converts colored-pencil line drawings into binary information (l's and O's).
Then the computer, acting in the
role of an "engraver", puts together
an image composed of mic roscopic
spots, as commanded by the recognized colored-pencil codes.
Film
transparencies, each will millions
of the square spots, are created
on the optical scanner/plotter.
These transparencies are used for
making the lithographic plates for
the color printing presses.
The system was developed for the
Uni ted States Army Engineer Topographic Laboratories, Fort Belvoir,
Virginia.
It is described in detail in a paper entitled "ComputerAided Mapping: A Total System Approach," written by Dr. P. June Min
and Donald R. Thompson of IBM's
Systems Development Laboratory. A
copy of the paper is available from
IBM Corporation,Systems Develop-
ment Division, Neighborhood Rd.,
Kingston, N.Y. 12401.
METALLURGIST DESCRIBES
COMPUTER-MICROSCOPE
COMBINATION
An unusual experimental configuration consisting of a scanning
electron microscope (SEM) coupled
to an IBM System/360 Model 44 computer has been described by IBM
metallurgist Richard Pyle speaking
at the American Chemical Society's
161st National Spring Meeting. The
equi pment provides "on-Ii neff or immediate chemical analysis of the
specimen under observation. Where
the conventional light optical microscope can magnify up to 2,000
times, the SEM can make an object
appear more than 50,000 its real
size~
"Digital computers operate from
digi tal inputs," Mr. Pyle explained.
"Substituting adigital scanner for
an analog scanner eliminates several roadblocks. The computer is no
longer hindered from its task of
collecting and analyzing data from
the SEM.
On-the-spot analysis is
possible."
The paper, titled: "An On-Line
Scanning Electron Microscope/Pseudo
Electron Microprobe System," is coauthored by IBM's Poughkeepsie development
laboratory engineers,
Robert B. Togan, Timothy C. Hartmann, and Mark L. Shulman.
MISCELLANEOUS
JAPAN NOW EXPORTS
TECHNICAL KNOW HOW
Hitach~ Ltd., Tokyo, Japan, has
announced the availability of 513
of their U.S. patents for licensee
arrangements.
Selected from over
20,000 patents granted to the company, they cover eqUipment in several fields, including communications, electronic components, and
electronic computers.
Examples of the Hitachi patents
in the electronic computer category
include: a system for curve plotting including interpolation; an
electrical device for compensating
a digital execution time in hybrid
computer systems;
a pen-tracking
system in cathode ray tube display
equipment; and an opto-electronic
input and output memory device.
Further information may be obtained by contacting Licensing Operations, Hitachi America, Ltd., 437
Madison Ave., New York, N.Y. 10022.
COMPUTERS and AUTOMATION for May, 1971
NEW PRODUCTS AND SERVICES
DESCRIPTION
FOR MORE INFORMATION
For users whose computer requirements range from mediumscale operations to those combining large data bases,
remote computing, and extensive remote and local terminal networks / CYBER 70® family's four systems are
software compatiblej and users may link with CDC's
CYBERNET network whenever desirable
A custom OEM version of the MAC Jr. minicomputer / can
effect cost savings to 25% for many minicomputer users /
provides expansion from 2 to 28K words of 16-bit memory
within computer chassis / memory cycle is 1 useCj add
time is 2 usec / I/O structure expandable / selection
of standard peripherals / several options
Will accommodate 17 fifteen-inch printed circuit boards
in 1O~" of rack space / dual power supply has been added
to standard 800/1200 power supply / Jumbo Nova 1200 can
be upgraded to a Jumbo Nova 800 by simple exchange of
plug-in memory and CPU boards / to be shown at SJCC
General purpose computer combining intermediate-scale
DP capability with small-system economy / runs under
DOSj provides opportunity for using COBOL, FORTRAN and
RPG II / two main-storage sizes: 24,576 bytes and 32,7ffi
bytes / basic cycle time of CUP is 0.75 usec
Aerospace computer for nearly any environment / adaptable to wide variety of applications by changing or substituting memory and input/output modules /basic M-341
includes digital bus card, parallel-type central processor card, core memory several I/O modules and power
supply / uses include general testing and process control
Custom designed to user specifications / 16-, 12-, or
8-bit word lengths / asynchronous machine with typical
add times in 400- to 500-nsec range / application areas
include educational, medical, machine tool,manufacturing
Business-oriented mini with 100-200K bits (48 bitword
length) of magnetic drum memory / 2.5 msec average access time / direct memory access, accepts decimal as well
as binary arithmetic, interface system can handle to
15 input and 16 o~tput devices / two versions: Standard,
2048 words and Expanded, 4096 words
Multi-purpose aerospace digital computer, modular design / single data and control bus interconnects all
modules to standard interface / major elements include
central processor module, memory modules, I/O module
with built-in-test-equipment submodule and power supply module ! through-put varies wi th memory module used
For scientific and engineering applications / up to
131,072 words of core memory~ 900-nsec memory cycle
time, and 12 independent ports to memory / CPU is
word-oriented (32 bits plus parity) and also is addressable in 8-bit bytes, halfwords and doublewords /
compatible wi th Sigma 5 and 9 and all XDS peripherals
General purpose minicomputer / includes 4096 words of
core memory, a party line I/O bus, direct memory access,
power supply, automatic priority interrupt, a front panel
console wi th lock /620/i compatible, software and peripherals / low-cost ($5,400) / to be shown at SJCC
Control Data Corp.
8100 34th Ave. So.
Minneapolis, MN 55420
Attn: Kent R. Nichols
A computer-controlled tape library retrieval systemj
also performs administrative functions of record-keeping /includes one or more tape vault modules, one or
more shuttles, one minicomputer console / operates in
OS or DOS / First model, LCS-5, for the UNIVAC 1108
and IBM System/360-65 which typically have libraries
ranging ,from 2500 to 40,000 tapes
Advanced Digital Systems, Inc.
146 w. Main St.
Mohawk, NY 13407
Attn: Norbert Andres
NAME/MODEL NO.
Digital
CDC CYBER 70 family
Flexible MAC
Jumbo Nova 1200 and 800
IBM System/360 Model 22
Magic 341
PDP-16 Functional
Computer
PPC-50
SKC-2000
Xerox Sigma 8
Varian 620/L
Lockheed Elec tronics Co., Inc.
Data Products Division
6201 E. Randolph St.
Los Angeles, CA 90040
Attn: Wm. W. Orrange
Data General Corp.
Route 9
Southboro, MA 01772
Attn: Dick Brown
IBM Corp., Data Proc'g.Div.
1133 Westchester Ave.
White Plains, NY 10604
Attn: R. F. Whalen
Delco Electronics
Div. of General Motors Corp.
Milwaukee, WI 53201
Digital Equipment Corp.
146 Main St.
Maynard, MA 01754
Attn: Mark Nigberg
Peripheral Processor Co.
c/o GHB Advtg. Inc.
One East 42nd St.
New York, NY 10017
Singer-General Precision, Inc.
1150 McBride Ave.
Little Falls, NJ 07424
Attn: H. Magnuson
Xerox Data Systems
701 So. Aviation BlVd.
El Segundo, CA 90245
Attn: Chuck Ramsey
Varian
611 Hansen Way
Palo Alto, CA 94303
Special Purpose Systems
Automatic Tape Library
Control System
COMPUTERS and AUTOMATION for May, 1971
43
NAME/MODEL NO.
FOR MORE INFORMATION
DESCRIPTION
Memories
Ll07 Disc Memory
(dual-capacity)
Memory Sequence Control
1771 Disc Memery System
3-million bit storage capacity at 3600 RPM / 1800 RMP
version of same unit offers capacity of 3.7 Mbits /
applications include field-expandable extension of
minicomputer mainframe memory, communications terminal
storage. main or auxiliary memory for inventory control
Increases memory capab il i ty of indus trial robots / permi ts group of steps to be stored only once by robot / thereafter, robot will refer back to as often as required to complete program / may be added to Unima te indus trial robot as
field modification, essentially a matter of plugging it in
For Data General NOVA family of computers / head-pertrack system ranging in size from 65 thousand to 1 million
words on single disc / 17 msec average access time; wri te
protect swi tch / validity check in every data transfer
Librascope Div.
The Singer Co.
808 Western AVe.
Glendale, CA 91201
Unimation, Inc.
Shelter Rock Lane
Danbury, CT 06810
Data Disc Inc.
686 W. Maude Ave.
Sunnyvale, CA 94086
Attn: Ron Troxell
Software
BASIC-l
Business Planning Language (BPL)
CBISLEARN
EDP Auditor
Extended BASIC
Compiler/Interpreter
'MX03016', generalized
format print module
1401 SUPER/SIM
PRINT-FAST
SNOFLAKE Program Writing System
STENCIL
Telecommunications Programming System (TPS)
44
Gives NCR Century 100 system on-line scientific-problem
solving capability / two configurations: dedicated version, accommodating to 15 data terminals and entirely
devoted to BASIC-I; dual installation accommodates to
10 terminals, using 32K memorywith only 16K for BASIC-l
For developing and analyzing financial reports / uses
standard accounting procedures to create and evaluate
company's financial reports / easy-to-use language
Computer assisted instruction (CAl) system for use of
instructors not proficient in computers or programming /
written in APL; operates under APL PLUS system of Scientific Time Sharing Corp. / implemented ini tially to
work with multiple choice and true-false material
For auditing computer-based systems / written in Assembly Language / run s on IBM 360 DOS, OS, and RCA Spectra computers, TDOS and DOS / can be used by any audi tor
regardless of EDP experience
For small to medium scale time-sharing systems / permits multiple users to execute BASIC programs in fully
interactive time-share environment / design is modular /
may be tailored to particular requirements
An RPG-type program utilizing simple control cards to
furnish lists and/or other formatted output from a
single input card, tape or disk file / written in BAL
for IBM 360/0S and DOS; core requirement, 20K
Simulator which executes 1401 programs on S/360 under
the control of PC~ OS/MF~ OS/MVT / spooled output reduces operator intervention / supports all standard
1401 features / requires maximum of 80K core storage,
including 1401 program
For providing maximum overlap of processing and printing to yield maximum printer speed for Assembly Language and COBOL programs / runs on any TOS/DOS IBM 360
system / no modifications to system required
A function-oriented COBOL program writing system / user
need not be programmer / available functions include
report wri ting, label printing, file rebuilding and reformatting
A parameter driven universal printing routine,first of
a series designated FORMWARE / includes card, tape, or disk
input; printed or spooled output / self-relocating, as small
a parti tion as lOKI applications include personal ized letters, mailing labels, formatted dumps / ini tial version, 360/DOS
Lets minicomputers serve as front-end processor for IBM
System/360 and 370 / cons is ts of three modules - a Communications Processor Program; a 360/370-resident communications access method for non-emulatory preprocessor applications; anda Communication Program Generation
facility for creating new Communications Processor
Programs without writing assembly language code for
minicomputer or removing it from communications processing service
The National Cash Register Co.
Main & K Sts.
Dayton, OH 45409
Intrnatl. Timesharing Corp.
4620 W. 77th St.
Minneapolis, MN 55435
Attn: R. E. Greiling
CBIS Informations Systems
6355 Topanga Canyon Blvd.
Suite 307
Woodland Hills, CA 91364
Cull inane Corp.
One Boston Place
Boston, MA 02108
Polymorphic Corp.
460 California Ave.
Palo Alto, CA 94306
Computer Systems
29230 NW Evergreen
Hillsboro, OR 97123
Hygain Technologies, Inc.
65 Whitney St.
Westport, CT 06880
Atlantic Software Inc.
312 Lafayette Bldg.
Fifth and Chestnut Sts.
Philadelphia, PA 19106
Software Engineering, Inc.
143 Newland St.
Norton, MA 02766
Attn: Stephen N. Mills
Data Systems Auditors, Inc.
325 Chestnut St.
Philadelphia, PA 19106
Attn: S. L. Carroll
PHI Computer Services, Inc.
800 Massachusetts Ave.
Arlington, MA 02174
Attn: John Pryke
COMPUTERS and AUTOMATION for May, 1971
NAME/MODEL NO.
DESCRIPTION
FOR MORE INFORMATION
Peripheral Equipment
CMC 761 Printer
DEC terminal I Display
Terminal
DICOMED 36 Image Display
GTU-l Interactive Data\
Input System
IBM 3735 Programmable
Buffered Terminal
Model ADAC 1200
Coupler/Modem
Multiplexor Input/Output
Processor (MIOP)
OMR-650 Desk-top Mark
Scanner
OPTRIX Unlimited Document Reader
Voice Response System,
Model 7050
Westinghouse 1600 Computer Terminal
For CMC KeyProcessingCD Systems / on-line or off-line /
speeds from average 356 Ipm to over 1000 Ipm / multiplecopy printout / 64-character, EBCDIC / compatible wi th
all industry-standard tape codes
Low-cost replacement of standard Teletypes on any computer using ASCII code /72 characters/line; 12-inch
screen, capacity of 20 lines / transmission speeds of
110, 150, 300 baud; options 600,1200, and 2400 baud
Offers both high resolution and ability to display four
images simul taneously / converts digi tal image information to pictorial form, using array of over 4 million
points / incorporates "random posi tion" capabili ty
For use as either on-line or off-line data entry and-or
retrieval uni t / includes custom-designed keyboard, microprocessor, video display system, and 2 magnetic tape
cassette recorder-reproducers
Stores over 20,000 bytes of information on one recording
track / data can be wri tten and retrieved at rate of a million bytes per second / 3 disk tracks store data; 2 hold
terminal's program / keyboard data entry / data may be
stored for later transmission, unattended
Main application expected to be with CRT terminals and
printers operating in 30 cps to 120 cps speed range /
maximum telephone handset data transmission rate of
1200 baud / transmission is asynchronous
A front-end proces sor for SYSTEMS 72 / uses only one
memory cycle per data transfer; can handle I/O operations up to one million bytes per second /basic MIOP
consists of 16 channels (expandable to 64), each dedicated to a standard com uter or s stems
eri heral
Scans punched cards; marked forms up to 8~" x 11"
accepts mark scan forms without modification / user
may mark to 4,000 positions on 8~" x 11" sheet / uses:
as data terminal with associated teleprinter; connected to off-line data collection device; or, on-line
to large scale CPU or minicomputer
Senses printed or written information on a document,
regardless of ink used, at rates of 30 documents per
second! ignores over-stamps, cancellation marks / sorts
checks wi tti. less than one error in 50,000/ high speed
writing detection applications
Converts any telephone to computer terminal/stores
virtually unlimited vocabulary / interfaces directly
and remotely wi th most general purpose computers / telephone data access lines expand up to 256 if desired
Self-contained, interactive CRT display for direct users
or original equipment manufacturers /built-in interfacing allows plug-for-plug teletypewriter replacement or
operation at data rates up to 9600 baud / functions
synchronously or asynchronously / data processing,
time-sharing and process control applications
Computer Machinery Corp.
2231 Barrington Ave.
Los Angeles, CA 90064
Attn: Ruth Wishner
Digital Equipment Corp.
146 Main St.
Maynard, MA 01754
Attn: Edgar Geithner
DICOMED Corp.
7600 Parklawn Ave.
Minneapolis, MN 55435
Attn: Stephen D. Posey
Hypertech Corporation
7343 West Wilson Ave.
Harwood Hgts., IL 60656
IBM Corporation
Monterey and Cottle Roads
San Jose, CA 95111
Attn: B. J. Beers
Anderson Jacobson, Inc.
1065 Morse Ave.
Sunnyvale, CA 94086
SYSTEMS
6901 W. Sunrise Blvd.
Ft. Lauderdale, FL 33313
Attn: Ted Swift
Decision Inc.
5601 College Ave.
Oakland, CA 94618
Attn: Bob Woodward
Optrix Unlimited Corp.
21 Spencer St.
Stoneham, MA 02180
Attn: Harold J. Weber
Marketing Dept.
Phonplex Corp.
789 Park Ave.
Huntington, NY 11743
Canadian Westinghouse Co. Ltd.
Information Dept.
P.O. Box 510
Hamilton, Ontario, Canada
Computer-Related Services
AUTOMED
Commercialoan Monitor
A medical information system being offered initially
in Illinois, Ohio, Indiana, Michigan / enables physicians to obtain instant access to patients' medical
records by a visual display terminal connected to a
Univac computer system in Cleveland / other benefi ts
include partial or total automated billing, completing
standard medical insurance forms, automated appointment routines, patient classification, providing
medical literature in hard-copy form, etc.
Provides a complete on-line computerized service for
banks /only equipment requirement is typewriter-like
terminal which connects to BankCom's computer by telephone / service enables banks to analyze commercial
lo~n experience through over 35 separate reports or
combinations of reports for increased management con-
Medical Data Systems Corporation (MDS)
c/o Sperry Rand Corp.
Univac Division
P.O. Box 500
Blue Bell, PA 19422
Attn: M. M. Maynard
Bank Computer Network Corp.
333 N. Michigan Ave.
Chicago, IL 60601
trol / operation easily learned in an hour's time
COMPUTERS and AUTOMATION for May, 1971
47
NEW CONTRACTS
Burroughs Corp., Detroit,
Mich.
General Services Administration, Washington, D.C.
Redcor Corporation, Woodland
Hills, Calif.
Transamerica Computer Co.,
San Francisco, Calif.
Computing and Software, Inc.
(C&S) , Los Angeles, Calif.
National Aeronautics and Space
Administration (NASA)
NASA's Goddard Space Flight
Center, Greenbelt, Md.
Honeywell Information Systems,
Phoenix, Ariz.
Internal Revenue Service
Philco-Ford Corp., Communications and Technical Services
(C&TS) Div., Willow Grove, Pa.
U.S. Air Force Air Materiel
Area, Tinker AFB, Okla.
Qantel Corp., Hayward, Calif.
Computer Usage Co., Inc.,
Greenwich, Conn.
TRW Controls Corp., Houston,
Texas
Xerox Data Systems,
El Segundo, Calif.
Centronics Data Computer
Corp., Hudson, N.H.
Core, Ltd., London, England
F&M Systems Co., Dallas, Texas
Port of Seattle, Seattle, Wash.
Ampex Corp., Redwood City,
Calif.
Los Angeles County, Calif.
Universal Business Machines,
Columbia, S.C.
United States Postal Service
Univac Division of Sperry Rand
Corp., Blue Bell, Pa.
National Gypsum Co., Buffalo,
N.¥.
Honeywell Inc., Information
Systems Div., Framingham,
Mass.
The Boeing Company, Seattle,
Wash.
PRC Information Sciences Co.,
Los Angeles, Calif.
Naval Electronics Laboratory
Center, San Diego, Calif.
Digital Resources Corp., Hybrid
Systems Div., Houston, Texas
Volkswagenwerk AG, Wolfsburg,
West Germany
Computer Complex, Inc.,
Houston, Texas
National Aeronautics and Space
Administration Manned Spacecraft Center Houston Texas
Department of Defense, Defense
Supply Agency, Washington, D.C.
Potter Instrument Co., Inc.,
Plainview, N.Y.
Leasco Information Products,
Inc., Bethesda, Md.
U.S. Office of Education, Educational Resources Information
Center (ERIC), Washington, D.C.
Electronics Associates Ltd.,
Burgess Hill, England
Bri tish Nuclear Design and Construction Ltd., Whetstone,
Leicestershire, England
Computer Communications, Inc.,
Culver City, Calif.
Midwest Stock Exchange Service
Corp.
48
Forty-three (43) B3500 computer systems for
the Navy's Uniform Automatic Data Processing System (UADPS) to be installed at 22
Naval locations in the U.S. and overseas;
systems will be used for inventory control,
movement of material from supply points to
Naval fleet and shore establishments, and
for financial control
KeyLogic systems, communication systems,
and data acquisition systems for lease
to Redcor customers
A one-year facility management contract
with options for two additional one-year
extensions; C&S will operate, maintain and
manage NASA's Slidell computer complex at
the Michoud Assembly Facility, Slidell, La.
Existing contract option renewal for one
year; C&S supports variety of DP services
at Goddard
Installation of direct data entry equipment
at three new IRS regional tax processing
centers (Tenn., Calif •• and N.Y.)
Engineering, furnishing and installing a
telecommunications network (code-named,
Scopecomm) for U.S. forces in Europe; it
will link centers in the Federal Republic
of Germany, Belgium, and the. Uni ted Kingdom
Purchase of Qantel business computer systems to be marketed through CUC's offices
Xerox Sigma 3, Sigma 5 and Sigma 7 computers which will be used in company's electrical utilities energy management systems
and pipeline control systems
Purchase of at least 2,000 Centronics model
101 line printers, and related spare parts
and accessories, over 3-year period; Core
will have rights to market the 101 on the
European continent and in United Kingdom
Design, manufacture, installation of a computer-based Central Control and Communications Facility for the Seattle-Tacoma International Airport
Second contract for equipment and engineering that will prepare previously ordered
television records system to be extended
for use by any law enforcement and justice
agency in the county
Production of 50 letter-sorting machines;
firm will provide the sorters, plus spare
parts and chairs
A UNIVAC 1106 computer system, the first
step in establishing corporate computer
and nationwide wire communications center
Three computers to be used in the "bras sboard" phase (a development less detailed
than a prototype) of the Air Force's Airborne Warning and Control System (AWACS).
Completion of development and documentation
for the Message Processing and Distribution
System (MPDS), a communications system for
the USS Chester W. Nimitz
Three SS-IOO Analog/Hybrid computers to be
used in expanding research and development
activities
Computer timesharing services; includes
data reduction of information received
durin A 0110 missions
Lease (with option to purchase of computer
peripheral equipment, valued at about $5
million,for use on IBM computers
Operation of the ERIC Document Reproduction
Service; firm will reproduce and market information products to the education community -- primarily on microfiche
An EAI 8800 analog/hybrid computer to supplement BNDC's 1500 amplifier. 6 console
analog known as 'Saturn' for nuclear reactor
and power station design problems
A nationwide computerized hihg speed message
switching and brokerage data proces sing network
$30.6 million
$7 million
(approximate)
$6+ million
$6.1 million
(approximate)
$6 million
$5.9 million
$5 million
$5 million
$3.7 million
$3 million
$2.1 mi 11 ion
$1. 2 mill ion
(approximate)
$1. 2 mi 11 ion
(approximate)
$1+ million
$816,000
$755,000+
$50,000
COMPUTERS and AUTOMATION for May, 1971
NEW INSlALLAliONS
Burroughs B2500 system
Burger King Corp., Miami, Fla.
Gilbert Associates, Inc., Reading,
pa.
Meister Brau, Inc., Chicago, Ill.
Burroughs B3500 system
Control Industries Inc., Kansas
City, Mo.
Pittsburgh National Bank, Pittsburgh, Pat
Control Data 1700 system
3M Company, Minneapolis, Minn.
Control Data 3200 system
V. Vranas and Co., Athens, Greece
Control Data 3300 system
Innsbruck University, Innsbruck,
Austria
Control Data 6400 system
NASA's Langley Research Center,
Hampton, Va.
University of Aarhus, Aarhus,
Denmark
Data General Nova 1200 system
Bullock's, Los Angeles, Calif.
Datacraft
University of Wisconsin, Engineering Computing Laboratory, Madison,
Wis.
DC 6024/3 system
Honeywell Model 635 system
U.S. Naval Academy, Annapolis, Md.
IBM System/3 Model 6
Gingiss Formalwear, Inc., Chicago,
Ill.
IBM System/3 Model 10
Columbus Heating and Ventilation
Co •• Columbus. Ohio
Forest E. Olson, Inc., Van Nuys,
Calif.
Nicholson Terminal & Dock Co"
River Rouge, Mich.
Lockheed MAC 16
NCR Century 100 system
NCR Century 200 system
Univac 1106 system
Pee Dee State Bank, Timmonsville,
S.C.
Colonial Pipeline Co., Plainfield,
N.J.
Crown Construction Co., Johnstown, Pat
Good Samaritan Medical Center,
Zanesville. Ohio
First Computer Services, a subsidiary of the First National Bank,
Ruston, La.
Kansas State Bank & Trust Co.,
Wichita. Kans.
Commercial and Industrial Computer
Services, Johannesburg, South Africa
University of Freiburg, West Germany
COMPUTERS and AUTOMATION for May, 1971
General accounting jobs and business applications,
including payroll, profit and loss statements,
sales analysis and forecasting, product mix analysis
(system valued at more than $560,000)
General accounting, cash flow forecasting, general
ledger, manpower scheduling, and others
(system valued at more than $460,000)
General accounting, inventory accounting, production scheduling and market reporting
(system valued at more than $346,000)
Improving and expanding services to over sixty
business firms in 500-mile area; primary markets
served include· direct mail, bank processing and
commercial type applications
(system valued at more than $484.000)
Savings accounts, Christmas Club and check credit
accounting, correspondent bank checking accounts
and sixty other banking programs
(system valued at more than $1 million)
Research and development projects; system connected
to laboratory instrumentation units -- applications
include chemical component analysis, spectroscopic,
electrical and radiation data analysis
Training students in computer programming and computer usage, and providing a data center for its
own requirements and for external clients
Centralized computer services for the university's
science, engineering, economics, philosophy and
computer science departments
Use as the scheduling element for the Center's
Computer Complex which includes 3 CDC 6600s and a
previously installed 6400; new 6400 will enable
optimal utilization of the systems
University administration, faculty research, student training, and regional data center applications
(system valued at $1.5 million)
Directing a point of sales system (American Regitel
Corporation) in Bullock's new Northridge store; one
Regitel system already is in use in the downtown
Los Angeles Bullock's
Handling expanded work load built up over a tenyear period using older equipment; Computing Laboratory serves 2400 undergraduates, 800 graduate
students as well as staff and faculty
Increasing computer resources; Academy expects to
double to more than 80 the number of courses a midshipman may take that require use of the computer
Helping to assure proper size and style is available
in headquarters and franchise stores in eight states
as well as for general ledger, payroll. invoicing
Computerized job costing system; comprehensive
accounting system planned
Real estate listings including as many as 1500
items of information about features of each piece
of property
Financial control functions including payroll needs
(in cash daily) of some 1200 stevedores; also extensive job cost reporting
Demand deposit accounting~ aged trial balance and
customer service applications
Reading meters, calculating and printing delivery
tickets on ten pipeline systems; also will guard
against pipeline leakage
Design problems encountered in development of
shopping centers and enclosed shopping malls
Controlling rising costs; over 100 programs in use;
each of 30 different departments are monitored
Maintaining a Central Information File for bank's
customers; also offers data processing services to
area public utilities, hospitals, schools, and commercial establishments
Automating all banking departments with NCR's Central Information File software package
On-line invoicing, stock updating, consumer credit
and general accounting in batch, remote batch and
time sharing modes of operation; also performs
service bureau work for the government
A wide range of scientific and engineering applications as well as training of students in computer
science
(system valued at $1.4 million)
49
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 t.ime 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, confirm, or comment on any figures.
Our census seeks to include all digital computers manufactured anywhere. We invite all manufacturers located anywhere to submit information for this census. We invite all our readers to submit information that would help make these figures as accurate and complete as
possible.
The following abbreviations apply:
(A) -- authoritative figures, derived essentially from information
sent by the manufacturer directly to COMPUTERS AND
AUTOHATION
C
figure is combined in a total
(D)
acknowledgment is given to DP Focus, Marlboro, Mass., for
their help in estimating many of these figures
E
figure estimated by COMPUTERS AND AUTOMATION
(N)
manufacturer refuses to give any figures on number of installations or of orders, and refuses to comment in any
way on those numbers stated here
(R) -- figures derived all or in part from information released
indirectly by the manufacturer, or from reports by other
sources likely to be informed
(Sl
sale only, and sale (not rental) price is stated
X -- no longer in production
information not obtained at press time
Part I of the Monthly Computer Census contains reports for United
States manufacturers. Part II contains reports for manufacturers
outside of the united States. The two parts are published in alternate months.
SUMMARY AS OF APRIL 15, 1971
NAME OF
NAME OF
MANUFACTURER
Part 1. United States Manufacturers
Autonetics
Anaheim, Calif.
(R) (1/69)
Bailey Heter Co.
Wickliffe, Ohio
(A) (4/71)
Bunker-Ramo Corp.
Westlake Village, Calif.
(A)
(3/71)
Burroughs
Detroi t, Mich.
(N)
(1/69-5/69)
Computer Automation, Inc.
Newport Beach, Calif. (12/70)
Control Data Corp.
Minneapolis, Minn.
(A)
(R)
(9/70)
COHPUTER
RECOHP II
RECOMP III
Bailey 750
Bailey 755
Bailey 756
Bailey 855
BR-130
BR-133
BR-230
BR-300
BR-330
BR-340
BR-1018
205
220
B100/B500
B2500
B3500
B5500
B6500
B7500
B8500
208/808
216/816
G15
G20
LGP-21
LGP-30
RPC4000
636/136/046 Series
160/8090 Series
924/924-A
1604/A/B
1700
3100/3150
3200
3300
3400
3500
3600
3800
6400/6500
6600
6800
7600
DATE OF
FIRST
INSTALLATION
11/58
6/60
11/61
2/65
4/68
10/61
5/64
8/63
3/59
12/60
12/63
6/71
1/54
10/58
7/65
2/67
5/67
3/63
2/68
4/69
8/67
6/68
3/69
7/55
4/61
12/62
9/56
1/61
5/60
8/61
1/60
5/66
5/64
5/64
9/65
11/64
8/68
6/23
2/66
8/64
8/64
6/67
12/68
AVERAGE OR RANGE
OF MONTHLY RENTAL
$ (000)
2.5
40-250
(S)
200-600
(S)
60-400
(S)
100-1000
(S)
2.0
2.4
2.7
3.0
4.0
7.0
23.0
(S)
4.6
14 0
2.8-9.0
5.0
14.0
23.5
33.0
44.0
200.0
5.0
(S)
8.0
(S)
1.6
15.5
0.7
1.3
1.9
NUMBER OF INSTALLATIONS
Outside
In
In
U.S.A.
World
U.S.A.
30
o
32
6
13
o
5
o
30
X
35
6
18
8
6
17
160
79
15
18
19
19
o
1
143
157
o
X
X
X
X
X
X
25-38
28-31
52-57
44
65-74
4
NUMBER OF
UNFILLED
ORDERS
12
18
o
o
o
7
13
2.1-14.0
11.0
45.0
3.8
10-16
13.0
20-38
18.0
25.0
52.0
53.0
58.0
115.0
130.0
235.0
27-40
30-33
X
X
64-69
62
72-81
4
117
190
o
1
150
170
295
20
165
322
75
29
610
29
59
106-180
83-110
55-60
200
20
15
39
20
85
85
1
8
60
13
130
215
x
X
X
X
X
X
X
X
o
C
C
C
C
C
C
C
C
C
C
C
Total:
160 E
Data General Corp.
Southboro, Mass.
(A)
Datacraft Corp.
Ft. Lauderdale, Fla.
Digiac Corp.
Plainview, N.Y.
(A)
(4/71)
(A)
(4/71)
(3/71)
Digital Computer Controls, Inc.
Fairfield, ~.J. (A) (3/71)
Digi tal Equipment Corp.
Maynard, Mass.
(A)
(2/71)
50
NOVA
SUPERNOVA
NOVA 1200
NOVA 800
SUPERNOVA SC
6024/1
6024/3
6024/5
Digiac 3060
Digiac 3080
Digiac 3080C
D-112
2/69
5/70
12/71
3/71
6/71
5/69
2/70
10/71
1/70
12/64
10/67
8/70
8.0
9.6
5.4
6.9
11.9
54-200
33-200
16-50
9.0
19.5
25.0
10.0
PDP-l
PDP-4
PDP-5
PDP-6
PDP-7
11/60
8/62
9/63
10/64
11/64
3.4
1.7
0.9
10.0
1.3
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
813
102
100
11
34
o
11
37
43
o
o
o
o
30
16
7
45
48
40
90
C
C
5
o
10
C
C
50
300
50
45
100
23
160
X
X
X
X
X
COMPUTERS and A"UTOMATION for May, 1971
NAME OF
MANUFACTURER
Digital Equipment Corp. (Cont'd.)
NAME OF
COMPUTER
PDP-8
PDP-8/1
PDP-8/S
PDP-8/L
PDP-9
PDP-9L
PDP-I0
PDP-11
PDP-12
PDP-IS
LINC-8
DATE OF
FIRST
INSTALLATION
4/65
3/68
9/66
11/68
12/66
11/68
12/67
3/70
9/69
2/16
9/66
AVERAGE OR RANGE
OF MONTHLY RENTAL
$ (000)
0.5
0.4
0.3
1.1
8.0
10.5
17.0
(S)
NUMBER OF INSTALLATIONS
In
Outside
In
U.S.A.
U.S.A.
World
1440
C
C
3698
C
C
1024
C
C
3902
C
C
436
C
C
48
C
C
145
C
C
C
C
546
C
C
475
6
C
15
C
C
142
NUMBER OF
UNFILLED
ORDERS
C
C
C
C
C
C
C
C
C
C
C
Total:
1350 E
Electronic Associates Inc.
Long Branch, N.J. (A) (4/71)
EMR Compu ter
Minneapolis, Minn.
(A)
(2/71)
640
8400
EMR 6020
EMR 6040
EMR 6050
EMR 6070
EMR 6130
EMR 6135
EMR 6155
4/67
7/67
4/65
7/65
2/66
10/66
8/67
1.2
12.0
5.4
6.6
9.0
15.0
5.0
2.6
95
19
55
6
140
25
C
C
C
C
C
C
C
C
C
C
Total:
107
Total:
15 E
12
General Automation, Inc.
SPC-12
1/68
Anaheim, Calif.
SPC-16
5/70
(R) (6/70)
System 18/30
7/69
General ~E71~e~ct~r-l~'c--------------------~G~E~-~P7A~C~30~1~0~--------~5~/~7~0-------------2~.0~------------~----------~0------------~----------~---West Lynn, Mass.
Process Control Computers
(A)
(12/70)
Hewlett Packard
Cupertino, Calif.
(A) (4/71)
Honeywell Information Systems
Wellesley Hills, Mass.
(A) (2/71)
GE-PAC 4010
GE-PAC 4020
GE-PAC 4040
GE-PAC 4050
GE-PAC 4060
2114A, 2114B
2115A
2116A, 2116B, 2116C
G58
GI0 SA
GI05B
GI05RTS
Gl15
G120
G130
G205
G210
G215
G225
G235
G245
G255 T/S
G265 T/S
G275 T/S
G405
G410 T/S
G415
G420 T/S
G425
G430 T/S
G435
G440 T/S
G615
G625
G635
G655
H-110
H-115
H-120
H-125
H-200
H-400
H-800
H-1200
H-1250
H-1400
H-1800
H-2200
H-3200
H-4200
H-8200
DDP-24
DDP-1l6
DDP-124
DDP-224
DDP-316
DDP-416
DDP-516
H1l2
H632
H1602
H1642
H1644
H1646
H1648
H1648A
COMPUTERS and AUTOMATION for May, 1971
10/70
2/67
8/64
12/66
6/65
10/68
11/67
11/66
5/70
6/69
6/69
7/69
4/66
3/69
12/68
6/64
7/60
9/63
4/61
4/64
11/68
10/67
10/65
11/68
2/68
11/69
5/64
6/67
6/64
6/69
9/65
7/69
3/68
4/65
5/65
12/70
8/68
6/70
1/66
12/67
3/64
12/61
12/60
2/66
7/68
1/64
1/64
1/66
2/70
8/68
12/68
5/63
4/65
3/66
3/65
6/69
6.0
6.0
3.0
7.0
2.0
0.25
0.41
0.6
1.0
1. 3
1.4
1.2
2.2
2.9
4.5
2.9
16.0
6.0
8.0
12.0
13.0
17.0
20.0
23.0
6.8
11.0
7.3
23.0
9.6
17.0
14.0
25.0
32.0
43.0
47.0
80.0
2.7
3.5
4.8
7.0
7.5
10.5
30.0
9.8
12.0
14.0
50.0
18.0
24.0
32.5
50.0
2.65
0.9
2.2
3.5
0.6
2
172
45
23
18
0
49
20
9/66
10/69
12/68
1.2
3.2
255
30
960
370
1075
86
73
325
185
10
20
185
22
20
14
90
250
250
60
450
350
900
75
12
11/68
12.0
20
200-400
420-680
11
35
15
145
40-60
1
15
17
15-20
45-60
15-30
o
10-40
170-300
50-100
20
70-100
20-30
6
15
125
20
18
10
16
59
X
X
X
620-1080
11
35
16
160
57-77
15-20
60-90
10
15-45
240-400
70-130
26
26
23-43
23
20-40
180
30
800
150
800
46
58
230
130
2
221
65
24
20
1060
319
940
75
160
220
275
40
15
90
55
6
5
60
2
X
X
X
X
X
51
NAME OF
MANUFACTURER
IBM
White Plains, N.Y.
(N)
(D)
(1/69-5/69)
Interdata
Oceanport, N. J •
(A) (4/71)
NCR
Dayton, ohio
(R)
(2/71)
Phi1co
Willow Grove, Pa.
(N) (1/69)
RCA
Cherry Hill, N.J.
(N)
(5/69)
Raytheon
Santa Ana, Calif.
(A)
(4/71)
Scientific Control Corp.
Dallas, Tex.
(A)
(6/70)
Standard Computer Corp.
Los Ange les, Cali f.
(N) (12/70)
Systems Engineering Laboratories
Ft. Lauderdale, Fla.
(A)
(6/70)
UNIVAC Div. of Sperry Rand
New York, N.Y.
52
DATE OF
NAME OF
FIRST
INSTALLATION
COMPUTER
System/3 Model 6
3/71
System/3 Model 10
1/70
System/7
11/71
305
12/57
650
10/67
1130
2/66
1401
9/60
1401-G
5/64
1401-H
6/67
1410
11/61
1440
4/63
1460
10/63
1620 1, 11
9/60
1800
1/66
7010
10/63
7030
5/61
704
12/55
7040
6/63
7044
6/63
705
11/55
7020,
3/60
7074
3/60
7080
8/61
7090
11/59
7094-1
9/62
7094-11
4/64
360/20
12/65
360/25
1/68
360/30
5/65
360/40
4/65
360/44
7/66
360/50
8/65
360/65
11/65
360/67
10/66
360/75
2/66
12/69
360/85
360/90
11/67
370/135
5/72
370/145
9/71
370/155
2/71
370/165
5/71
360/195
4/71
Model 1
12/70
Model 3
5/67
Model 4
8/68
Model
11/70
Model 15
1/69
304
1/60
310
5/61
315
5/62
315 RMC
9/65
390
5/61
500
10/65
Century 100
9/68
Century 200
6/69
1000
6/63
200-210,211
10/58
2000-212
1/63
301
2/61
501
6/59
601
11/62
3301
7/64
Spectra 70/15
9/65
Spectra 70/25
9/65
Spectra 70/35
1/67
Spectra 70/45
11/65
Spectra 70/46
Spectra 70/55
11/66
250
12/60
440
3/64
520
10/65
703
10/67
704
3/70
706
5/69
650
5/66
655
10/66
660
10/65
670
5/66
4700
4/69
DCT-'132
5/69
IC 4000
12/68
IC 6000
5/67
IC 7000
8/70
810
9/65
810A
8/66
810B
9/68
840
11/65
840A
8/66
840MP
1/68
Systems 86
I & II
3/51 & 11/57
III
8/62
AVERAGE OR RANGE
OF MONTHLY RENTAL
$ (000)
1.0
1.1
0.35 and up
3.6
4.8
1.5
5.4
2.3
1.3
17.0
4.1
10.0
4.1
5.1
26.0
160.0
32.0
25.0
36.5
38.0
27.0
35.0
60.0
63.5
75.0
83.0
2.7
5.1
10.3
19.3
11.8
29.1
57.2
133.8
66.9
150.3
(S)
15
18
1227
1836
450
140
116
1174
63
186
148
14
40
50
2580
2210
420
180
156
1690
194
285
415
67
4
12
35
28
18
10
44
13
4
10
6
4690
27
13
26
3276
4
3144
498
13
109
31
4
5075
1260
65
480
175
9
14
o
5
14.4
23.3
48.0
98.7
232.0
3.7
8.5
10.5
20.0
14.0
2.5
8.7
12.0
1.9
1.5
2.7
7.5
7.0
40.0
52.0
7.0
14.0-18.0
14.0-35.0
17.0-35.0
4.3
6.6
9.2
22.5
33.5
34.0
1.2
3.6
3.2
12.8
9.8
19.0
0.5
2.1
2.1
2.7
1.8
0.9
9.0
16.0
17.0
1.1
0.9
1.2
1.5
1.5
2.0
10.0
25.0
21.0
NUMBER OF INSTALLATIONS
Outside
In
In
U.S.A.
U.S.A.
World
45
N/A
200
25
40
15
8
400
150
950
1100
1100
350
16
16
12
140-290
22-50
o
10
55
200
300
40
64
17
172
(S)
64
55
23
137
41
(S)
5
o
8
300
45
500
1800
300
90
700
195
1450
2900
1400
440
70
X
90
30
13
X
X
X
X
X
100-130
1
1-5
35-60
18-25
20-50
21-55
o
(S)
41
21
13
70
15
6
14
10
7966
4
8219
1758
78
589
206
13
17
240-420
23-51
o
2
24-60
90-110
68-70
65-100
84-180
11
115
20
26
55
68
3807
4046
870
320
272
2864
257
471
563
81
5
13
o
o
100
5
24
NUMBER OF
UNFILLED
ORDERS
20
1
31
16
14
19
45
25-65
l25-170
86-95
85-150
105-235
1
12
175
20
27
203
80
69
23
137
41
1
19
45
X
X
X
2
32
X
o
X
4
23
9
24
o
III
5
75
24
216
76
3
o
36
31
o
38
31
o
o
o
23
25
2
31
X
32
26
X
X
2
X
X
COMPUTERS and AUTOMATION for May, 1971
NAME OF
MANUFACTURER
UNIVAC (Cont I d.)
(2/71)
(A)
Varian Data Machines
Newport Beach, Calif.
(A) (4/71)
Xerox Data Systems
El Segundo, Calif.
lR)
(2/71)
DATE OF
NAME OF
FIRST
COMPUTER
INSTALLATION
File Computers
8/56
Solid-State 80 I,ll,
90, I, II, & Step
8/58
418
6/63
490 Series
12/61
1004
2/63
1005
4/66
1050
9/63
1100 Series (except
1107, 1l08)
12/50
1107
10/62
1108
9/65
9200
6/67
9300
9/67
9400
5/69
LARC
5/60
620
1l/65
620i
6/67
R-620i
4/69
520i
10/68
520/DC
12/69
620/f
1l/70
620/L
6/71
XDS-92
4/65
XDS-910
8/62
XDS-920
9/62
XDS-925
12/64
XDS-930
6/64
XDS-940
4/66
XDS-9300
11/64
Sigma 2
12/66
Sigma 3
12/69
Sigma 5
8/67
Sigma 6
6/70
Sigma 7
12/66
Sigma 9
BOOKS (Continued from page 35)
Arno Press, Inc., 330 Madison Ave., New York, NY
10002 / 1967, hardbound, 311 pp., $?
Hearings, and statements from some dozen
persons, on the positive and negative aspects of
storing personal data in computers.
Enrick, Norbert L. / Decision-Oriented Statistics /
Brandon Systems Press, Inc., 1101 State Rd.,
Princeton, NJ 08540 / 1970, hardbound, 216 pp.,
$8.95.
This book presents statistical analysis methods to help managers analyze relationships, evaluate courses of action, calculate risks, and arrive
at decisions. The methods described include the
use of statistical analysis for application in
today's computer information systems. Some other
subjects covered are computer simulation systems,
network analysis, strategy under risk and uncertainty, and decision-making and the computer.
Katzan, Harry~ Jr. / APL Programming and Computer
Techniques / Van Nostrand Reinhold Company,
450 West 33 St., New York, NY 10001 / 1970,
hardbound, 329 pp., $12.00.
This book is intended to give a thorough foundation in the APL language (of K. Iverson) and a terminal system, and to provide a thorough introduction to computer techniques for scientists, engineers, business analysts, and managers.
Cuadra, Carlos A., editor, and 17 authors / Annual
Review of Information Science and Technology
(Volume 5) / Encyclopedia Britannica, Inc., 425
North Michigan Ave., Chicago, IL 60611 / 1970,
hardbound, 476 pp., $17.50
Gives 13 significant reports on individual
laboratorY,or experimental pilot programs of the past year in the field of information
sciences.
research~
COMPUTERS and AUTOMATION for May, 1971
AVERAGE OR RANGE
OF MONTHLY RENTAL
$ (000)
15.0
NUMBER OF INSTALLATIONS
In
Outside
In
U.S.A.
U.S.A.
World
13
8.0
11.0
30.0
1.9
2.4
8.5
210
76
75
1502
637
138
35.0
57.0
68.0
1.5
3.4
7.0
135.0
9
8
87
1051
387
8
36
11
628
299
62
114
822
49
0
0
0.4
1.6
0.5
1.5
2.0
2.9
3.0
3.4
14.0
8.5
1.8
2.0
6.0
12.0
12.0
35.0
2
7-10
5-12
1
14
0
NUMBER OF
UNFILLED
ORDERS
X
112
86
2130
936
200
9
11
56
175
144
2
75
1300
50
150
25
7
10-60
150-170
93-120
20
159
28-35
21-25
60-110
10
15-40
10-15
0
6-18
12 62
157-180
98-132
21
173
28-35
22-26
70-125
10
21-58
24-35
5-9
29-44
X
20
35
20
90
10
E
E
E
E
E
X
X
75
850
550
60
E
E
E
E
X
400
30
330
25
125
65
Profit, A. / Structure et Technologie des Ordinateurs / Armand Colin, 103, Blvd. Saint-Michel,
Paris 5, France / 1970, hardbound, in French.
648 pp., $?
This is a manual in French intended for students and engineers who wish to learn about the
internal organization of machines and what may be
loosely termed hardware. It covers two fundamental
aspects: the description of circuits and the composition of computers, and a study of the organization of these different elements for the construction of systems.
The 22 chapters include "Algorithmic Structures and the Organization of Calculators", "Logic
Systems of Semi-Conductors", "General Organization
of a Calculator", "Unity of Command", and "The
Structure of the Computer". There is an index and
a bibliography.
The author is a professor at L'Ecole Superieure
d'Electricite and has had some experience in the
laboratories at the French National Center of Studies of Telecommunications. The manual is u2sed
on a course taught by the author.
ADVERTISING INDEX
Following is the index of advertisements. Each item
contains: name and address of the advertiser / page
number where the advertisement appears / name of
agency, i f any
ACADEMIC PRESS, 111 Fifth Ave., New York, N.Y. 10003
/ Page 56 / Flamm Advertising
ASSOCIATION FOR COMPUTING MACHINERY, 1133 Avenue of
the Americas, New York, N.Y. 10036 / Page 2 /
Corporate Presence, Inc.
COMPUTERS AND AUTOMATION, 815 Washington St., Newtonville, Mass. 02160 / Page 3
JOHN WILEY & SONS, INC., PUBLISHERS, 605 Third Ave.,
New York, N.Y. 10016 / Page 35 / Ed McLean & Co.
XEBEC SYSTEMS, INC., 918 N. Rengstorff, Mountain
View, California / Page 35/ Moorhead, Moran & Lanig
53
The Golden Trumpet
STOP EXPRESSING POLITICAL POLEMICS
John Culleton
Data Processing Manager
Penberthy Mfg. Co.
P.O. Box 112
Prophetstown, III. 61277
Really, if you want subscription renewals, devote yourselves to computers and automation. Most
of us are sick unto death of political polemics on
every subject except the technical topics we want
to learn about. I don't go to political social science journals to keep up with the state of the computational art and I certainly don't go to a technical'EDP journal to hear about assassinations and
court trials. The slim technical content is largely superficial. I get several better technical
journals free. Advertisements are important too,
and your ads are nearly nonexistent.
complaint that I have had about an increasing number of articles in your magazine. The issue is one
of relevance.
In the March 1971 issue, for example, you've
published Kingman Brewster again. There isn't even
the pretense of a computer connection in this one.
A technical journal should apply itself to the field
it professes to cover. Editorializing should be
labelled as such, and segregated. Special purpose
articles, unrelated to the technical area should be
the exception, not the rule. No matter how much
hair spli tting you do in your "Responses", all your
non-computer articles seem to be selling the same
political ideas.
If you want to go into competi tion wi th "poli tical comment" type of magazines, by all means do so,
but be honest about it. Stop masquerading as a
computer journal.
DON'T BLAME PRESIDENT NIXON
I. From James E. 'Crow
207 Mulberry Road
Newark, Del. 19711
I am sorry to read in the January issue of C&A
that you blame President Nixon for "an economic
depression and storm-sweeping through American business", etc. Apparently you haver,' t done your
homework. Nixon found runaway inflation and had
been attempting to control it. He didn't initiate
Vietnam or "Vietnamization". Check the figures
to see how he is winding down the Vietnam war, in
both men and dollars. The cutbacks for defense
have hurt. What do you want? Don't let your
emotions carry you away and detract from the good
work you are doing.
2. From H. Irvin Smith
54746 Yarnall
Houston, Tex. 77035
By inference you blame President Nixon and the
United States' involvement'in Southeast Asia for
the declining advertising volume in Computers and
Automation. This is somewhat surprising since one
would think that the so-called "Military-Industrial
Establishment" would be a much larger user of computers than would be the various radical gr~ups and
liberal social causes you so ardently champlon.
Have you considered the possibility that many
advertisers may prefer not to be identified with
your left-wing editorial content? (Just as they
don't advertise in far-right political journals.)
This also may be a factor in the eight-to-fifteen
percent decline in your paid circulation over the
past year. I, for example, plan not to renew my
company's subscription.
MASQUERADING AS A COMPUTER JOURNAL
Patrick M. Cooney
RD6
Carmel, NY 10512
In the March 1971 issue of Computers and
Automation, Dr. Schwartz raised one point in his
refutation of R. Sprague's article "The Assassination •••• ", with which I heartily agree. It's a
54
SURPRISE, DELIGHT, AND SUPPORT FOR YOUR
VIEWS AND CONCERNS
John R. Macleod
130 Rosevale Valley Rd.
Toronto 5, Ontario, Canada
I
It was only recently that several delayed issues
of "Computers and Automation" (May and November,
1970) crossed my desk, a publication I must admit I
had not previously read, or, for that matter, even
heard of. I am in the academic field, teaching Operations Research at Ryerson Poly technical Institute
in Toronto, and this not-having-read-it previously
was very surprising, to say the least. It may be
that we have only recently begun receiving it.
In any event, imagine my surprise to find social
issues presented, discussed, editorialized, etc. -in a computer journal of all placesl I was even beginning to think that I was alone among my esteemed
colleagues in being concerned over social matters.
To be interested in people might even seem to be
mutually exclusive of being interested in things,
the latter being usually all too true of people in
technical fields. I therefore read with a great
deal of interest your articles on John F. Kennedy's
assassination, the plight of Clark Squire, etc. and
all of the comments on these matters. It certainly
is a refreshing change from the usual right-wing or
not-giving-a-damn non-involved academic publications
which are so widely circulated. I am even tempted
to wonder if such enlightenment is more of a common
denominator among systems people, a field which implies the broad interdisciplinary approach that
only intelligence can cope with successfully. It
sometimes does seem to me that the really competent
people in this field do have a way above average intelligence and social concern. This is a field for
further research, as the academic says.
This letter therefore is to express my surprise,
delight, and continued support for your policies and
views and concerns, and to provide a personal paton-the-back for someone who dares to make use of his
opportunity to express a fair, balanced, enlightened,
honest opinion in social matters in print, an all
too rare occurrence in our unnecessarily complex
society. Keep at itl You've got one more avid supporter and promoter in 'Canada now, that maybe you
didn't realize you had.
COMPUTERS and AUTOMATION for May, 1971
PUBLISHING COURAGE
Robert W. Carter, President
QICSYS Systems Inc.
128 West 58 St.
Bayonne, N. J. 01702
I am writing to express my feelings to you on the
generally excellent form and content of your publication, "Computers and Automation." I am currently in the position of wearing a number of "hats"
which call for varied readings as a user, teacher,
and manufacturer of computer science and technology.
As a result, my readings cover a very broad spectrum, and I can easily rate "Computers and Automation" as one of the best.
Please accept my best wishes for your continued
success and my admiration of your publishing courage.
OASIS OF TRUTH
Mrs. Grace P. Vale
St. Louis, Mo.
Computers and Automation is like an oasis of
truth amidst the mass media fiction about the political assassinations, and I would like to express
my appreciation to you for printing this information.
I have been having a great deal of trouble with
my mail ever since I started to subscribe to Mr.
Joachim Joesten's Truth Letter, one of the few other
sources of information on this subject. I would
greatly appreciate it if the circulation department
of Computers and Automation would give me some information about an order I placed with you .•.•
On Wednesday I received Computers and Automation,
and The Atlantic Monthly, and, of course,
read eGA almost straight through before even looking
at the others.
Harper~v
I especially enjoyed your answer to the attack
on Richard Sprague's article." The articles on
information or data processing, the book review,
and several others in the March issue were also
extremely interesting.
Since that March 17, 1961 when the world's first
mini (costing $120,000) was delivered, Digital
Equipment Corporation has installed over 10,000 more
to perform a host of applications throughout the
world. Today's mini is priced under $5,000 and has
more power than the $120,000 original.
The estimated $1.2 billion market by 1975 for
minis certainly implies many new and exciting developments.
THOMAS J. WATSON FELLOWSHIPS AWARDED
TO 70 STUDENTS FROM 34 COLLEGES
Robert O. Schulze, Executive Director
Thomas J. Watson Foundation
220 South Main St.
Providence, R. I. 02903
The Thomas J. Watson Foundation announced on
March 19 the award of 70 fellowship grants totaling
$438,000.
The grants will make possible a year of independent postgraduate travel and study abroad for
70 students from 34 colleges and universities in
the United States. Each of the 61 unmarried recipients will receive an award of $6,000 and each of
the nine married winners will receive $8,000.
The fellowships are awarded annually by the foundation, a charitable trust established by the late
Mrs. Thomas J. Watson, Sr., in memory of her husband, founder of IBM Corporation.
The winners were chosen from among graduating
seniors nominated by participating colleges. They
were selected primarily on the basis of their potential for creative leadership in their chosen fields,
although academic records and extracurricular activities were also taken into account. Each will
pursue a travel-study program devised by the fellow~
ship winner to increase his or her personal or career potential.
The fellows' travel-study programs will take
them to Europe, Africa. South America, Asia and
other areas, with several visiting more than one
continent.
Mark Nigberg
Digital Equipment Corp.
46 Main St.
Maynard, Mass. 01754
Their interests range from the study of medical
care in developing countries to the role of nonviolence in resolving conflicts. Richard A. Reines
of Tufts University, Medford, Mass., will study the
former in Africa and India; Michael W. Johnson of
Whitman College, Walla Walla, Wash., will investigate the latter in Europe, Africa and Asia.
With the advent of general-purpose digital computers 25 years ago, everyone thought of the giant
octopus reaching out to control the world. This
thinkinIT prevailed for the next 15 years. But on
March 17, just 10 years ago, a new product and with
it a new philosophy was introduced: the mini-computer.
The fellowship program was begun in 1968 by
the daughters and sons of the late Mr. and Mrs.
Watson, Sr.: Mrs. Walker G. Buckner, the late Mrs.
John N. Irwin II, IBM Chairman Thomas J. Watson,
Jr., and Arthur K. Watson, U.S. Ambassador to France
-- working with Dr. Robert 0. Schulze, executive
director of the foundation.
The mini was but an ant compared to the octopus.
But like the ant, it was dedicated to its work. It
would work on a smaller scal~ in a more simplified
operation. And like the ants, the minis would work
together, each doing its assignment and all together
successfully completing the job.
Mr. and Mrs. Watson, Sr. were long interested
in education and world affairs. For much of his
business career, Mr. Watson. Sr. was involved with
the Chamber of Commerce, and he vigorously supported the concept of "World Peace Through World Trade."
He was an early and strong supporter of the United
Nations. Both Mr. and Mrs. Watson traveled widely
in connection with the global business requirements
of IBM.
THE ANT AND THE OCTOPUS
As the ants greatly outnumber the octopus, the
day when minicomputers greatly outnumber all
others is imminent.
COMPUTERS and AUTOMATION for May, 1971
55
,,'
,.1
II
THE THEORY OF
LINEAR SYSTEMS
GENERAL DYNAMICAL
PROCESSES
by J. E. RUBIO, Department of Electrical Engineering,
Carnegie-Mellon University, Pittsburgh, Pennsylvania
A Volume of Electrical Science
Series Editors: HENRY BOOKER
and NICHOLAS DECLARIS
Presents the theory of linear systems in a form suitable
for study by first- and second-year graduate students
and reference by practicing engineers. The book
introduces the theory of linear spaces and the fundamentals of differential and difference systems with
finite-dimensional state spaces. Other topics treated
are basic descriptions, controllability and observability, synthesis, and stability. The final chapter discusses
the general theory of dynamical systems, including an
introduction to the subject of semi-groups of operators,
and applies this theory to some simple differential
systems with state spaces of infinite dimensions. An
appendix deals with computational procedures. May
1971, about 329 pp., $14.00.
by T. G. WINDEKNECHT, Michigan Technological
University, Houghton, Michigan
A Volume of Mathematics in Science and Engineering
Series Editor: RICHARD BELLMAN
The author uses set theory to develop a powerful
axiomatic approach to general systems theory. He
covers such topics as dynamic processes in general
and their classifications, basic notions of combination
and interconnection, time-evolution, the strong types
of causality, and the concept of state. Results are
obtained without introducing assumptions of linearity,
finiteness, or time-variance and without, for the most
part, distinguishing discrete-time from continuoustime processes. Complete proofs are given and a large
number of mathematical exercises are included. 1971,
192 pp., $9.50.
ITERATIVE SOLUTION OF
LARGE LINEAR SYSTEMS
by DAVID M. YOUNG, Center for Numerical Analysis,
University of Texas at Austin, Austin, Texas
A Volume of Computer Science and Applied Mathematics
Series Editor: WERNER RHEINBOLDT
Provides a systematic development of a substantial
portion of the theory of iterative methods for solving
large linear systems with sparce matrices such as often
arise in the numerical solution of elliptic partial differential equations by finite difference methods. The book
also treats the successive overrelaxation method (SOR
method) including several variants and related methods. Convergence properties of the various methods
are studied in terms of the spectral radii of the associated matrices as well as in terms of certain matrix
norms. June 1971, about 560 pp., in preparation.
MATHEMATICAL SOFTWARE
edited by JOHN R. RICE, Department of Computer
Science, Purdue University, West Lafayette,' Indiana
A Volume of ACM Monograph Series
Series Editor: ROBERT L. ASHENHURST
The scope and content of the 22 papers presented at a
symposium on Mathematical Software held at Purdue
University in April, 1970, cover a broad spectrum of
topics in mathematical software. There are three introductory chapters by the editor which trace the historical background, review and analyze the current
status and present remarks on the future of the field.
The monograph then discusses: comprehensiyesurveys and developments (arithmetic, libraries,nonnumerical software), critical evaluation of current software (manufacturer's elementary function libraries,
quadrature algorithms), new approaches and systems
(ordinary differential equations, mathematical programming), philosophical "think" pieces (arithmetic,
library construction), and specific research results
(random numerical generation, singular values of
matrices). June 1971, about 510 pp., in preparation.
ADVANCES IN COMPUTERS
Volume II
Guest Editor: MARSHALL C. YOVITS, Division of
Computer and Information Science, Ohio State University, Columbus, Ohio
CONTENTS: HARRY H. JOSSELSON: Automatic
Translation of Languages Since 1960: A Linguist's
View. D. M. JACKSON: Classification, Relevance, and
Information Retrieval. KLAUS W. OTTEN: Approaches
to the Machine Recognition of Conversational Speech.
DAVID R. HILL: Man-Machine Interaction Using
Speech. R. B. KIEBURTZ and E. E. NEWHALL: Balanced Magnetic Circuits for Logic and Memory Devices. ANTHONY DEBONS: Command and Control:
Technology and Social Impact. Author Index-Subject
Index.
1971, 428 pp., $18.50.
INTRODUCTION TO
PROBABILISTIC AUTOMATA
by AZARIA PAZ, Department of Computer Science
and Mathematics, Israel Institute of Technology, Haifa,
Israel
A Volume of Computer Science and Applied Mathematics
Series Editor: WERNER RHEINBOLDT
This book-the first published in English on the subject
-discusses both the practical and theoretical aspects
of probabilistic automata and sequential machines.
The first chapter, dealing with state theory from an
engineering standpoint, covers the synthesis of stochastic machines, state minimization, equivalence,
coverings, and input-output relations. The second
chapter is entirely devoted to nonhomogeneous Markov chains, which provide the mathematical model on
which stochastic automata are based. It offers the first
discussion of this important topic to appear in book
form. The final chapter presents a theory of formal
stochastic languages and events, including such topics
as closure properties, decision problems, and characterizations. May 1971, 248 pp., $13.00.
ACADEMIC PRESS \(!/
fnNEW YORK AND LONDON
111 FIFTH AVENUE, NEW YORK, N. Y. 10003
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 22:00:10-07:00 Create Date : 2009:03:24 22:00:10-07:00 Metadata Date : 2009:03:24 22:00:10-07:00 Format : application/pdf Document ID : uuid:de75a1f5-952b-4a20-a5e8-806cf5aaee10 Instance ID : uuid:f0b1cae5-1587-46f4-8148-d33dca3787f7 Page Layout : SinglePage Page Mode : UseNone Page Count : 52EXIF Metadata provided by EXIF.tools