197307

197307 197307

User Manual: 197307

Open the PDF directly: View PDF PDF.
Page Count: 52

Download197307
Open PDF In BrowserView PDF
a

SCIENCE & TECHNOl0GY

Ii 53ap

July, 1973
Vo l. 22, No . 7

CDI •• puters
and automation

and people

"PLEASE TIM TICKLE LANA"

Communication - Three-way: Chimpanzee,
Man, Computer / Larry B. Dendy, Ernest
von Glasersfeld, and others
Computer Privacy and Security / Howard
Campaigne and Lance J. Hoffman
Computer Programming Using Natural
Language - Part 2 / Edmund C. Berkeley,
Andy Langer, and Casper Otten
Applications of Computers in Correspondence
Schools / Major Charles Ehin
The Right That Guards All Other Rights
John Gardner
The American News Media and the
Assassination of President John F.
Kennedy -

Conclusion

Richard E. Sprague

5106049 01 0 P 7401

TECHNICAL SERVICES
SA JOSE PUSl LIBRARY
180
AN CARLOS ST
SA J E C

201721

~,

018
95113

IF YOU COULD PREVENT
JUST ONE IMPORTANT MISTAKE
BEFORE IT HAPPENED.-· •
HOW MUCH

WOULD THAT BE WORTH TO YOU - $100? - $1000?

more?

Our considered estimate is that 10 to 20% or more of the cost of operation of most businesses is the cost of
mistakes. (Just one foreseeable mistake that "Computers and Automation" made in 1970 has cost us $4000.)

WOULDN'T YOU AGREE

THAT SENSE, COMMON AND UNCOMMON,
OUGHT TO BE THE KEY TO PREVENTING MISTAKES?
In a number of the issues of "The Notebook on Common Sense, Elementary and- Advanced", we examine
systematically the prevention of mistakes, such as:
No. 15:

Preventing Mistakes from Failure to Understand

No. 23:

Preventing Mistakes from Forgetting

No. 38:

The Concepts of Feedback and Feedback Control

No. 41:

Preventing Mistakes from Unforeseen Hazards

)
)

- Volume 1, first
subscription year
- Volume 2, second
subscription year

Among the forthcoming issues of the Notebook in Volume 2 are:
- Preventing Mistakes from Camouflage
- Preventing Mistakes from Placidity
and we are planning at least 20 more issues in Volumes 2 to 4 under this general heading.

WHY NOT TRY THE NOTEBOOK ON COMMON SENSE?
GUARANTEE: (1) You may return (in 7 days) the
first batch of issues we send you,
REFUND, if not satisfactory. (2)
may cancel at any time, and you
a refund for the unmailed portion
Scription.

PAST ISSUES: As a new subscriber, you do not miss past

for FULL
Thereafter, you
will receive
of your sub-

issues. Every subscriber's subscription starts at Vol. 1,
No.1, and he eventually receives all issues. The past
issues are sent to him usually four at a time, every week
or two, until he has caught up, and thus he does not miss
important and interesting issues that never go out of date.

HOW CAN YOU LOSE?
- - - - - - - - - - - - - - - - - - - - (may be copied on any piece of paper) - - - - -

To:

- - - -

- - - - -

- - - -

- - - -

COMPUTERS AND AUTOMATION AND PEOPLE
815 Washington St., R12, Newtonville, Mass. 02160

(

) YES, please enter my subscription to the Notebook on Common Sense at $12 a year,
24 issues (newsletter style), and extras.
) Please send me (as FREE premiums for subscribing) the first six issues:
1. Right Answers - A Short Guide to Obtaining Them
2. The Empty Column
3. The Golden Trumpets of Yap Yap

) I enclose $

(

4. Strategy in Chess
5. The Barrels and the Elephant
6. The Argument of the Beard

) Please bill me

( ) Please bill my organization

NAME ______________________________________________ TITLE
ORGANIZATION

-----------------------------------------------------------------------------

ADDRESS __________________________________________________________________________________
SIGNATURE __________________________________________ Purchase Order No . _______________________
2

COMPUTERS and AUTOMATION for July. 1973

INVENTORY OF THE 36 ISSUES OF

- TITLES AND SUMMARIES

THE NOTEBOOK ON COMMON SENSE, FIRST YEAR
VOLUME 1

1. Right Answers - A Short Guide to Obtaining Them
A collection of 82 principles and maxims. Example:
"The moment you have worked out an answer, start
checking it - it probably isn't right."
2. The Empty Column
A parable about a symbol for zero, and the failure
to recognize the value of a good idea.
3. The Golden Trumpets of Yap Yap
4. Strategy in Chess
5. The Barrels and the Elephant
A discussion of truth vs. believability.
6. The Argument of the Beard
The accumulation of many small differences may
make a huge difference.
7. The Elephant and the Grassy Hillside
The concepts of the ordinary everyday world vs.
the pointer readings of exact science.
8. Ground Rules for Arguments
9. False Premises, Valid Reasoning, and True Conclusions
The fallacy of asserting that the premises must first
be correct in order that correct conclusions be
derived.
10. The Investigation of Common Sense
11. Principles of General Science and Proverbs
8 principles and 42 proverbs.
12. Common Sense - Questions for Consideration
13. Falling 1800 Feet Down a Mountain
The story of a skimobiler who fell 1/3 of a mile
down Mt. Washington, N.H., and was rescued the
next day; and how he used his common sense and
survived.
14. The Cu It of the Expert
15. Preventing Mistakes from Failure to Understand
Even though you do not understand the cause of
some trouble, you may still be able to deal with
it. The famous example of a cure for malaria.
16. The Stage of Maturity and Judgement
17. Doomsday in St. Pierre, Martinique - Common Sense
vs. Catastrophe
How 30,000 people refusing to apply their common
sense died from a volcanic eruption.
18. The History of the Doasyoulikes
19. Individuality in Human Beings
Their chemical natures are as widely varied as
their external features.
20. How to be Silly
71 recipes for being silly. Example: "Use twenty
words to say something when two will do."
21. The Three Earthworms
A parable about curiosity; and the importance of
making observations for oneself.
22. The Cochrans vs. Catastrophe
The history of Samuel Cochran, Jr., who ate some
vichyssoise soup.
23. Preventing Mistakes from Forgetting
24. What is Gammon Sense? An Operational Definition
A proposed definition of common sense not using
synonyms but using behavior that is observable.
25. The Subject of What is Generally True and Important Common Sense, Elementary and Advanced
26. Natural History, Patterns, and Common Sense
Some important techniques for observing.
27. Rationalizing and Common Sense
28. Opposition to New Ideas
Some of the common but foolish reasons for
opposing new ideas.
29. A Classification and Review of the Issues of Vol.
30. Index to Volume 1

COMPUTERS and AUTOMATION for July. 1973

VOLUME 2
31. Adding Years to Your Life Through Common Sense
32. The Number of Answers to a Problem
33. "Stupidity has a Knack of Getting Its Way"
34 and 35. Time, Sense, and Wisdom
36. Wisdom - An Operational Definition
.... 24 issues promised, 36 issues delivered, for good measure

Some Comments from Subscribers
bel ieve these to be the best, if not the most important,
reading that I have had this year.
- Harold J. Coate, EDP Manager, St. Joseph, Mo.
Your concept is brilliant, and a welcome antidote to much
which is passed off as useful knowledge these days. Keep
up the good work.
- Charles E. Abbe, Data Systems Analyst, Pasadena,
Calif.
Very good articles; something all managers should read.
- William Taylor, Vice President, Calgary, Alberta
As I am involved with systems work, I can always use one
of the issues to prove a point or teach a lesson.
- Edward K. Nellis, Director of Systems Development,
Pittsford, N.Y.
Thoroughly enjoy each issue.
- David Lichard, Data Processing Manager, Chicago, I II.
All are good and thought-provoking - which in itself
is worthwhile. Keep it up.
- Richard Marsh, Washington, D.C.
Especially like "Right Answers".
- Ralph E. Taylor, Manager of Research and Development, West Chester, Ohio
Your tendency to deal with practical applications is very
rewarding.
- Jeffrey L. Rosen, Programmer, Toronto, Canada
PAST ISSUES: As a new subscriber, you do not miss past issues. Every subscriber's subscription starts at Vol. 1, no.
1, and he eventually receives all issues. The past issues
are sent to him usually four at a time, every week or
two, until he has caught up, and thus he does not miss
important and interesting issues that never go out of date.
GUARANTEE: (1) You may return (in 7 days) the first batch
of issues we send you, for FULL REFUND, if not satisfactory. (2) Thereafter, you may cancel at any time, and
you will receive a refund for the unmailed portion of
you r su bscription. ~

WE WANT ONLY HAPPY AND SATISFIED SUBSCRIBERS.
- - - - - - - - - - (may be copied on any piece of paper) - - - - - - - - - - - - - - To: Computers and Automation and People
815 Washington St., R1, Newtonville, Mass. 02160
) YES, I would like to try the "Notebook on Common
Sense, Elementary and Advanced". Please enter my
subscription at $12 a year, 24 issues, newsletter style,
and extras. Please send me issues 1 to 6 as FREE
PREMIUMS for subscribing.
I enclose _ _ _ _ __

) Please bill me.

) Please bill my organization.
Name

Title _ _ _ _ _ _ _ _ __

Organization _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __
Address _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ ___
Signature ___________ Purch. Order No. _ _ __

3

Vol. 22, No.7
July, 1973
Editor

Edmund C. Berkeley

Assistant
Editors

Barbara L. Chaffee
Linda Ladd Lovett
NeilD. Macdonald
Thomas A. McLaughlin

Software
Editor

Stewart B. Nelson

Advertising
Director

Edmund C. Berkeley

Art Director

Ray W. Hass

Contributing
Editors

John Bennett
Moses M. Berlin
Andrew D. Booth
John W. Carr III
Ned Chapin
Leslie'Mezei
Bernhard W. Romberg
Ted Schoeters
Richard' E. Sprague

. Ed Burnett
Advisory
Committee James J. Cryan
Bernard Quint

Editorial
Offices

Advertising
Contact

Berkeley Enterprises, Inc.
815 Washington SL
Newtonville, Mass. 02160
617-332-5453

The Publisher
Berkeley Enterprises;' Inc.
815 Washington St.
Newtonville, Mass. 02160
617 -332-5453

'''Computers and Automation" is published monthlYI 12 issues per year, at 815
Washington St., Newtonville, Mass. 02160,
by Berkeley Enterprises, Inc, Pririted in
U.S.A. Second Class Postage paid at Boston,
Mass., and additional mailing points.
Subscription rates: United States, $9.50
for one year, $18.00 for two years. Canada:
add 50 cents a year for postage; foreign, add
$3.50 a year for postage.
NOTE: The above rates do not include
our publication liThe Computer Directory
and Buyers' Guide".
If you elect to receive "The Computer Directory and Buyers'
Guide," please add $9.00 per year to your
subscription rate.
Please' address all mail to:
Berkeley
EnterpriSes, Inc., 815 Washington St., Newtonville, Mass. 02160.
Postmaster: Please send all forms 3579
to Berkeley Enterprises, Inc., 815 Washington St., Newtonville, Mass. 02160.
© Copyright 1973, by Berkeley E nterprises, Inc.
Cha'nge 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

computers
and ·autornaCion
'and people
Computer Applications
7 Communication - Three-Way: Chimpanzee,· Man,
[NT A)
Computer r Part 1: Introduction / Part 2: A Computer
Mediates Communication with a'Chimpanzee
by Larry B. Dendy, Ernst von Glasersfeld,. and others,
University of Georgia, Athens, Ga.
How Lana, a chimpanzee, is using a language (called
"Yerkish") to express the meanings of. what she wants to say.
25 Applications of Computers in C9rre~pondence Schools
[T A]
by Major Charles Ehin, Maxwell AFB, .Ale:
How extensive and how sophisticated are the flJJpH{;Rtigm?
of computers'in correspondence schools?

. Computers and Privacy
12 Computer Privacy and Security
[NT A]
by Howard Campaigne and Lance J. Hoffman, Slippery Rock
State College, Po,
Because of new data pr~qessing technology it is becoming
feasible to access automqtj~ajjy \1M! quantities of information
about individual persons: whot §tJrt§ gf oontmls are reasonable
and are needed?

Computer Programming
18 Computer Programming Using Natural Language -Part 2
[T A]
by Edmund C.Berkeley, Andy Langer, nnfl Ca~per Otten,
Newtonville, Mass.
Input -' ordinary naturallangua£H~ ~xpressinga calculation;
Output - a precise computer program to p€trfgrm~glqJlatiol1;
two more specific simple examples; and a dialog on th@ y~:'
fulness of programming in ordinary natural language.
50 . Computer-Assisted Analysis and Documentation of
[NT G]
Computer Programs
Two books for persons seeking to understand working binary
programs having' operating instructions but no documentation
and no explanation.

The Profession of Information EnginCfr and the Pursuit of Truth
38 Unsettling, Oistu.rbing, Critical
Statement
policyJJV Cf)mpIJUtr~ and Auto l17atipq

[NT F]

of

31 The American News Media and th~A~$Cl~S~"itJon,Bf
~~1
President John F. Kennedy: Accessories After ·the Fijet
- Conclusion
by Richard E. Sprague,' Hartsdale, N.Y.
An examination of what happened in many important
American news organizations, to cover up and hide the
facts about how President John' F. Kennedy was actually
assassinated in. Dallas.

e.1

6 High-Level Crime and the High-Level Criminal
{NT E]
by EdmundC. Berkeley, Editor, and Thomas A. McLaughlin,
Assistant Editor, Computers and Automation
Probably the Equity Funding Life InsuranceCo. frauds
were made possible by computers; but a deeper trouble
is the attitudes of Americans towards fraud and crime.
COMPUTERS and AUTOMATION for July. 1973

The magazine of the design. applications, and implications of
information praco.9stng systems - and the pursuit of truth in.
trU~lI;, output, and processing, for the benefit of people.

The PrQfesslon

of Information Engineer and

the Pursuit of Truth (continued)
30 The Right That Guards All Other Rights
[NT G]
by John Gardm~rJ Chairman, Common Cause, Washington, D.C.
"If freedom of information is successfully stifled, all the
other freedoms can be snuffed out at leisure. What citiiens
don't know, they can't object to."

Computers, Science, and Common Sense

Hornet
[NT G]
by Edrnl.lnd C. Berkeley, author of Ride the East Wind:
Parables of Yesterday and Today
How far one may be from o!leaplng through a screen "only gfaw little thin wires between the inside and freedom"
- and what to do about it.

39 The Fly, the Spider, and the

3 The C&A Notebook on Common Sense, Elementary
and Advanced
Titles of the first 36 issues, and capsule summaries
of many.

[NT G]

2 Preventing Mistakes Before They Happen

[NT G]

Front Cover Picture
See the leading article, page 7 ...

Departments

Computer People
51 "Who's Who in Computers and Data Processing"
[NT G]
Th~late~t cumulative edition (the 5th edition published in
1971, containing over 15,000 capsule biographies) plus two
supplements (over 2000 more entries)

Computers and Puzzles
17 Numbles
by Neil Macdonald

[T C]

49 Problem Corner
by Walter Penney, COP

[T C]

Corrections
For changes in liThe Path to Championship Chess by Computer" by
Professor Donald Michie published in the January 1973 issue of
Computers and Automation, see page 36 of this issue (July).
I n the June issue, the footnote of page 36 requires correction. This
footnotf3 ~hQuld be corrected to read as follows:
Accessorie~ After the Fact is the title of a book by Silvia Meagher
published by Bobbs Merrill in 1967, accusing the Warren Commission
and the various government agencies of criminally concealing the
conspiracy to kill President John F. Kennedy. This article accuses
the national news media of the same crime.

40

30
46
47

44
45

Across the Editor's Desk Computing and Data
Processi ng .Newsletter
Advertising Index
Calendar of Coming Events
Monthly Computer Census
New Contracts
New Installations

Key
[A]
[C]
[E]
[F]
[G]
[NT]
[T]

-

Article
Monthly Column
Editorial
Forum
The Golden Trumpet
Not Technical
Technical

NOTICE

NOTICE: The official name of this magazine throughout 1973 is
Computers and Automation. We expect to change the name
officially as of January 1, 1974, to Computers and People.
During 1973 from time to time, unofficially, and irregularly,
we plan to use the name Computers and Automation and
People as a way of informing our subscribers and readers of
the intended change on January 1, 1974.
COMPUTERS and AUTOMATION for July, 1973

*D ON YOUR ADDRESS IMPRINT
MEANS THAT YOUR SUBSCRIPTION INCLUDES THE COMPUTER
DI RECTORY.
*N MEANS THAT
YOUR PRESENT SUBSCRIPTION
DOES NOT INCLUDE THE COMPUTER DIRECTORY.

5

EDITORIAL

High-Level Crime and the High-Level Criminal
The Wall St. Journal of April 24 published an article
"Equity Funding Scandal Grows, Involves $120 Million in
Nonexistent Assets" by Priscilla S. Meyer, in which the
following occurs:
Equity Funding Life Insurance Co. sold a package
that involved both life insurance and mutual fund shares.
The purchaser agreed to invest a certain amount of
money in a mutual fund. The shares he got were then
used as collateral for a loan from Equity Funding that
was used to pay the premium on the insurance policy.
. . . The company was selling the insurance policies for
cash to the various reinsurers.
Someone at the company decided that it would be
easy to create phony policyholders, sell them phony insurance, and peddle this to the reinsurers. . .. Not only
were the policies phony but so too was a gigantic
bunch of assets carried on the books of the parent Equity Funding Corp. These assets represented the loans
allegedly made to Equity Funding Life policyholders
.. , One source now says that about $77 million of
these loans are phony. ..,
The customers didn't exist. Their mutual fund shares
didn't exist. The funded loans didn't exist. The phony
customers' phony pledges of their phony fund shares to
buy phony insurance ultimately became numbers on a
computer tape, which then printed out phony assets for
Equity Funding Corp.'s phony books. ..,
In Parade Magazine for May 6, appears an article "How
to Bribe a Congressman" by the columnist Jack Anderson,
in which the following appears:
Lobbyist Cyrus T. Anderson, convicted of paying
Senator Daniel Brewster (Democrat, Maryland) "an unlawful gratuity" amounting to $14,500, told the court
that he funneled thousands of dollars, most of it in hard
cash, from corporations and unions to various politicians.
What kind of people came to him for cash: "Senators, Congressmen, candidates for President of the United States" said Lobbyist Anderson. He was scrupulous,
however, never to bribe anyone inside a federal building.
"It was my understanding," he explained to the court,
"that even legal political contributions can't be made in
a federal building, much less ones that weren't going to
be reported." Why do Congressmen want payment in
cash? "They prefer," replied Anderson, "not to make
public the sources of their incomes."
What are the reactions of most Americans to crime at
these high levels? There is no question that their reactions
are different to crime at the level of robbery of a store or
holding up a bank.
"The real weakness in the moral position of Americans,"
wrote the sociologist and Progressive scholar Edward A.
Ross in his book Sin and Society, 1907, "is not their attitude towards the plain criminal, but their attitude towards
the quasi-criminal." The latter, he said, escape the condemnation of the public because their sins do not share in
the Biblical simplicity of low-level crimes. The high-level
criminal threatens the purse of large impersonal groups, not
your purse and mine. Their kind of wrongdoing has evolved along lines that are parallel to the impersonal nature of
industrial society, and so is left untouched by traditional
6

morality. The high-level criminal, Ross said, is characterized not by his tendency to do evil in a personal way but
by his moral insensibility. "He is a buyer rather than a
practitioner of sin," Ross said, "and his middlemen spare
him unpleasant details."
Of course, high-level criminals are even more evil than
their low-level analogues. This kind of criminal is the executive of a life insurance corporation who helps sell phony
insurance policies to phony customers. He is an administrative aide who because of "loyalty to the boss" coversup the organizing and planning of: electronic eaves dropping; burglarizing; fake letters; false newspaper advertisements; manufactured "public" support for the bombing of
Hanoi and Haiphong; etc. He is a former attorney general
of the United States who when indicted says, ')'1 did not
steal any money". He is a former Assistant Secretary of
Defense who says, "Any government has a right to lie to
save itself".
The high-level criminal rejects the crude instruments of
street crime, but with recent advances of technology has
gathered together a more efficacious arsenal. His tools of
violence include: diverted campaign contributions carefully
collected in cash and "laundered" by passing them through
bank accounts in Mexico; the programs and operations of
the organization's computer; the interstaff memo and special "Gemstone" stationery; the offer of the directorship of
the Federal Bureau of Investigation to a judge in charge of
an important trial in which the government is prosecutor;
etc.
The high-level criminal understands the difference in
public esteem between those who do the burglarizing and
the implanting of electronic bugs, and those who merely
order it; he seeks to place bureaucratic distance between
himself and the evil committed under his auspices. He admits to no more at any stage of the news than he has to.
First, he calls what goes on a "bizarre incident". Later, he
calls the explanations he has previously given "inoperative".
Finally, he says he acted to conceal crime because of "the
interests of national security".
Dr. Samuel Johnson, English lexicographer, essayist and
poet, in 1775, said "Patriotism is the last refuge of a
scoundrel". Today's Madison Avenue translation of "patriotism" is "national security".
The antidote to high-level crime is enforcement of the
laws, the removal of criminals from power and authority,
and the indignation of honest and courageous Americans
who are resolved to have a better and a more honest socciety. We have had such a society in the past. It is time
to have it again.
And incidentally such a society is good business: it
costs much less and produces much more. It would give
more employment, less inflation, lower taxes, less fictitious
and more real security, etc., to every American. For it
costs a lot of money to have criminals in charge of a computer, a life insurance company, or a society.

~c.~ ~tl.//h~
Edmund C. Berkeley
Editor

Thomas A. McLaughlin
Assistant Editor
COMPUTERS and AUTOMATION for July, 1973

Communication -

Three-way:

Chimpanzee I Man I Computer
Part 1. Introduction
Larry B. Dendy
Public Relations Office
The University of Georgia
117 Terrell Hall
Athens, Ga. 30601

"The animal can 'read' in her language and make grammatically correct requests for
food and entertainment."

A chimpanzee named Lana at the Yerkes Regional
Primate Research Center is "talking" with the aid of
a computer and a new language created especially for
her.
The animal can "read" in her language and can
make grammatically correct requests for food and
entertainment. Later on the scientists hope to carry on a conversation with her.
Lana "speaks" by punching out sentences on a
special vertical keyboard. The requests are answered by a computer, which will grant them if the
sentence is gr&mmatically correct and refuse them
if Lana makes a grammatical mistake.
The research project is sponsored by the National
Institute for Child Development and is conducted
jointly by researchers from the University of Georgia, Georgia State University, and the Yerkes Center
at Emory University. Seven scientists from the
three institutes have been working on the project
for more than a year.
Communication

Dr. Ernst von Glasersfeld, a psychologist at the
Oniversity of Georgia, designed and implemented
Lana's special.language. He explained:
"We wan ted to find out if, in a -con trolled environment, chimpanzees, gorillas and orangutans can
be taught to communicate spontaneously through the
use of a language-like system for their own purposes. Through this we hope to find out something

2

6

ox o
o
3

7

4

8

5

9

Figure 1. The great majority of the lexigrams which Lana uses are
made up of the 9 single design elements shown here.

COMPUTERS and AUTOMATION for July. 1973

significant about the basic problems many children
have in learning their language."
Von Glasersfeld said the scientists were interested in two levels of communication. First, they
wanted to see if the apes could convey wants and
needs, for which they would receive material rewards. Then, they wanted to see if the apes could
be taught to communicate without the promise of
material rewards - to describe something, to "talk"
about something they did not necessarily want.
"If they can do that," said von Glasersfeld,
"then it might be possible to get them to ask questions. It might be possible to set up a facility
for man-to-ape and ape-to-man communication and maybe even ape-to-ape communication."
Several attempts to teach non-human primates to
speak English have failed because the animals' vocal
chords are not equipped for the modulations and
emissions necessary to produce English word sounds,
von Glasersfeld said.
However, two studies made during the past six
years .have proved beyond doubt that, when apes are
given an effective visual communication technique,
apes can express various requests and can learn some
of the essential parts of linguistic communication.
Elements 2,4,7

Orange

BOX

Red

$
~

. TO BITE

Elements 5,9

Red

B
WATER

NUT
Elements 1,3,8,9

Orange

~

ii

BLANKET
Elements 1,5,6

Elements 1,7,8

Blue

Elements 1,2,3,5

Blue

~

TO GROOM

Figure 2. Samples of Lana's lexigrams. The numbers indicate the
elements of the lexigram; the colour is ind icated in the right-hand
top corner, the "meaning" underneath.

7

The Language "Yerkish"

One important question von Glasersfeld and his
colleagues wanted to examine was what factors influence or restrict the apes' ability to acquire a
language. To make the study objective and errorfree, the scientists created a special language
called Yerkish, and designed and built a computercontrolled facility in which to teach it to apes.
The language Yerkish is built from nine simple
figures - a vertical line, an equals sign, a rectangle, a caret (or inverted v),a large open circle, a small black circle, a black triangle, a diamond-shaped figure, and a horizontal wavy line~
(See Figure 1) By superimposing these figures on
each other, the~cientists have created graphic
symbols that they call lexigrams, that stand for
various concepts.

lexigrams, She consistently picks out the ri~htones
in the right order to make a grammatically correct
reques t.
"Reading"

Further proof that she "reads"ot1rl Understands
the visual symbols :is shown when the chimp accidently hits the "please" key while hanging onto the
activating bar. If she notices the "please" symbol
on the screen, she does not strike the key again but
continues to add her request to it.
Also, if one of the scientists punches the first
two or three keys of a sentence, Lana can finish it
correctly.
The fact that she notices and understands the
visual symbols is important. This encourages the
scientists to hope that eventually they can ask Lana
questions and engage her in conversation.

For example, an open circle intersected by a
wavy line means "water". A vertical line intersecting a black circle enclosed by an open circle
is the symbol for "nut". (See Figure 2)
"Words"

The present computer system can handle 256 lexigrams or "words". So far the scientists have created 125 lexigrams. To avoid ambiguity, each lexigram
has only one specific meaning; this is unlike English, in which most words have more than one specific meaning.
The symbols are on a vertical keyboard (see
Figure 3) attached to a computer that has been programmed to deal with the language. Above the keyboard are seven small projection screens on which
the symbols indicated by pressing keys appear in
the sequence in which they are pressed.
In order to "talk," Lana must activate the system
by pulling a bar above the keyboard. She must punch
out the proper lexigrams in the proper sequence to
make her request and then end the sentence with a
period symbol.
"Grammar"

The computer will accept only sentences that are
grammatically correct in Yerkish. If Lana makes her
request correctly, the computer automatically triggers a dispenser that gives her what she has asked
for. If she makes a mistake, the computer sounds an
error buzzer, erases all symbols from the screens,
and Lana must begin again.
An example of the sentence structure she must
follow is, "Please machine give me a piece of banana". Ins tead of "banana," she can indicate "orange, candy, raisin, water, milk" or other foods.
She can also ask for music, a picture, a short movie, a toy, and for the window to be opened.
The computer makes a totally objective judgment
of Lana's linguistic performance. This eliminates
the possibility that human prejudices or sympathies
or errors in regard to Lana will interfere with
training her or judging her progress or her competence.
The computer records all Lana's efforts for later
study by the scientists.
The researchers know that Lana has not simply
memorized the proper sequence to punch on the keyboard because, when they change the position of the
8

Figure 3. Photographs of the·25 keys of the keyboard which Lana uses
are here shown with actual intervening 'spaces removed. The board at this
time displays 25 lexigrams (see tabular statement below). Some lexigrams
shown are: no. 5, PERIOD, no design, solid yellow; no. 9, PIECE, elements 1, 5, 6,8, orange .ground; no. 20, WINDOW, elements 2, 7, 8, orange ground; no. 23, PLEASE, black arrow, yellow ground. (The word
lexigram here is used to include "sentence modifier", but sentence modifiers are NOT composed of the regular design elements.)

•

PROJECTOR

MILK

BLANKET

SHELLEY

TOY

TO
GROm!

NO

PIECE

MUSIC

OF

MACHINE

MOVIE

M&M

CHOW

TIM

JUICE

DIFFERENT

MILK

WINDOW

TO
MAKE

WHAT

PLEASE

OPEN

TO
GIVE

The upper right frame contains the sign for a period, which Lana uses
to finish her sentences.
.

"Knowledge"

Lana now knows between 35 and 40 lexigrams. (See
Tables 1 and 2)
New ones are being introduced.
When she can produce a certain number of ~entences
correctly and can learn a new lexigra~ and use it
correctly in a new sentence, "then she will have
achieved productivity - the ability to form a
correct sentence with new words," said von GlasersCOMPUTERS and AUTOMATION for July, 1973

reilt. "1'hat is one of the requirements of language
competence •.

Table 2
SENTENCES OF LANA, MAY, 1973

Table 1
LEXIGRAMS OF LANA, MAY, 1973

Following is Lana's actual vocabulary, i.e., the
lexfgrams she is using correctly as of May,1973:
APpLE!
MOVIE
BANANA
BLANKET
CHOW
COME (activity)
DOOR
GIVE (acti vi ty)
GROOM (activity)

INTO

jUfCE
LANA (in instrUctions
to her)
MACHINE
MAKE (activity)

M&M
MILK

MUSIC
NO
OF
OPEN
PIECE
PLEASE
ROOM
SHELLEY (name of experimenter)
TICKLE (acti vi ty)
'rIM (name of experimenter)
TO

TOY
WATER
WINDOW

"If she responds correctly· to
tinued, "then we will be able to
conceptu~l mechanism conforms to
mechanism humans use and express

1. Sentences actually used by Lana:
PLEASE MACHINE GIVE BLANKET
JUICE
MGM

MILK
TOY
WATER
PLEASE MACHINE GIVE PIECE OF APPLE
CHOW
BANANA
PLEASE TIM
TICKLE LANA
SHEL.LEY
PLEASE TIM
GROOM LANA
SHELLEY
PLEASE TIM
COME INTO ROOM
SHELLEY
PLEASE MACHINE MAKE MUSIC
MOVIE

questions," he contest how far her
the conceptual
in language."

Other members of the group working with von Glasersfeld are: Dr. Duane M. Rumbaugh, former vicedirector of the Yerkes Center and now head of the
psychology department at Georgia State University;
Dr. Josephine Brown, professor of psychology at
Georgia State; Piero Pisani of the University of

PLEASE MACHINE MAKE WINDOW OPEN
2. Sentences she responds to:
PLEASE LANA COME TO DOOR
PLEASE LANA GROOM TIM
SHELLEY
Georgia's Computer Center; Prof. Harold Warner and
graduate students Charles Bell and Timothy Gill of
the Yerkes Center.

Part 2. A Computer Mediates
Communication with a Chimpanzee
Ernest von Glasersfeld
Harold Warner
Pier Paolo Pisani
Duane M. Rumbaugh
Timothy V. Gill
Charles Bell

Yerkes Regional Primate Research
Center of Emory University
1364 Clifton Rd., N.E.
Atlanta, Ga. 30322

"Until very recently it was considered a self-evident proof of man's uniqueness
among the fauna of this planet that he had developed and was using the
sophisticated communication system we call language."

A computer has recently been employed to explore
the possible linquistic capabilities of an ape (Rumbaugh et ~., in press). Sponsored by NICHHD at the
Yerkes Regi onal Primate Research Center of Emory University' a team of specialists in the areas of psychology, primatology, psycholinguistics, computer systems,
and bio-medical engineering, from the Yerkes Primate
Cente~ Georgia State University, and the University
of Georgia, is studying language-relevant behaviors
of a two-year old infant chimpanzee called Lana.
The purpose of the project is twofold. First,
to develop an ape-machine-man communication system
for the exploration of the linguistic potential of
apes; second, to gain, in the course of that exploration, insights into the processes of concept formation and language acquisition that may prove useful
in the design of remedial programs for retarded
children and, generally. for basic linguistic incompetencies in humans.
COMPUTERS and AUTOMATION for July. 1973

This article primarily intends to describe the
computer-controlled test facility that was designed
for the above purpose.
Background

Until very recently it was considered a selfevident proof of man's uniqueness among the fauna
of this planet that he had developed and was using
the sophisticated communication system we call language. As a result, it was assumed that other
species, because they had not developed such a system, must also be incapable of using it. Indeed,
Chomsky (1968), drawing on the Cartesian dogma of
mind-matter dichotomy and the consequent split between man and all other living organisms, went so
far as to assert that language was to be considered
dependent on man's species-specific genetic endowment.
9

During the last five years this belief has been
severely shaken by two studies. The Gardners (1971)
succeeded in teaching Washoe, an infant chimpanzee,
enough American Sign Language (the language traditionally used by deaf mutes) so that Washoe can now
communicate requests and even some spontaneous comments in this medium. Independently, David Premack
(1971) trained Sarah, again a female chimp, in the
use of an artificial language made up of coloured
plastic symbols that can be arranged into sentencestrings on a magnetic board. According to his reports, Sarah is now able, among other things, to answer questions, to handle negation, and even to respond correctly to instructions that involve a condi tion ("if ... then").
Both these studies show that chimpanzees, though
they do not get beyond a relatively simple signalling system when left to their own devices, in the
wild, can indeed learn to use a nonvocal languagelike means of communication and can acquire at
least some parts of what linguists call "linguistic
competence" (Ploog and Melnechuk, 1971). How much
of this they can acquire, however, is still an open
question.
Objective Records

Two things seem essential in an attempt to throw
more light upon this question of competence. The
syntax of the language to be acquired by the experimental subject must be rigidly defined before
the subject's performance can be assessed; and the
assessment of the subject's performance must be made
in an objective manner. 3 We therefore decided to
take the evaluation of grammaticality or well-formedness of the subject's linguistic output out'of the
hands of a human observer and to let a patently unemotional and unbiased computer decide whether or
not the linguistic performance of the subject was
correct according to the pre-established rules of
the language.
Using a computer to monitor and evaluate the subject's entire linguistic production at once provides
another advantage: the computer has no difficulty in
keeping a complete record of all interactions for
subsequent analYSis and investigation. Human observers hardly ever achieve such absolute completeness in their records, and, in long-term studies,
they tend to constitute a forbidding cost.
As it functions at present, our communication
facility comprises four main elements.
Yerkish, an Artificial Language

Yerkish is a visual language using graphic wordsymbols we call "Lexigrams". The computer implementation can hold a vocabulary of 250 lexigrams,
fifty of which are actually being used at the moment. All lexigrams are two-dimensional designs
consisting of one, two, three, or four constant design elements. There are nine design elements and
three colours (yielding seven discrete colour combinations) chosen for their discriminability and
cOmbinability (see Fig. 1).

(von Glasersfeld, 1970). Correlational grammar differs from traditional grammars in that (a) its wordclasses are based on a semantic or conceptual classification, not on morphological surface characteristics of words; and (b) its syntax is based on a
detailed specification of relational concepts and
their applicability to items of the conceptual wordclasses, rather than on the very generic syntactic
functions or connectives of traditional grammars.
Thus, instead of the usual word-class "noun," Yerkish has at present 15 lexigram-classes of items that
can function as "subject" in a sentence; instead of
the usual word-class "verb," there are ten lexigramclasses of items that designate different activities; and instead of the tradi tional "subj ect-verb"
function, Yerkish now has nine "correlators" (out of
a total of 29) repre~enting conceptually different
"actor-activity" relations.
As a result of this concept-sensitive snytax,
word combinations such as "(the) banana eats" or
"Lana drinks (a) banana," which according to traditional grammar would be syntactically correct and
only semantically deviant, are inadmissible according to the correlational syntax of Yerkish.
In the present implementation of the system (operational since December 1972) sentences are limited to a length of seven lexigrams and an end-sign
equivalent to a periodo A sentence is defined as
the string of lexigrams between a start signal
(triggered by the subject's grasping and hanging on
to a bar above the keyboard) and the period signo
Keyboard and Feedback Projectors

A keyboard of 125 keys arranged in five panels of
25 each (two such panels are operational at present)
is the language input device of the system. Each
key bears one lexigram. Depressing a key activates
the corresponding lexigram entry in the computer's
dictionary and transfers the semantic and syntactic
representation of the designated concept to the
workspace of the parser. Simultaneously the computer sends a code signal representing the design
elements of which the particular lexigram is composed to an external decoding device which, in turn,
activates one of seven projectors above the keyboard
to reconstruct the visual lexigram by projecting the
required design elements (one superimposed on the
other) onto its ground glass screeno The lexigram
of the first key that is depressed appears in the
first projector from the left, the second lexigram
in the second, and so on, so that the lexigrams
forming the sentence appear in their linear sequential order from left to right above the keyboard.
Above the feedback projectors there is a second
row of similar devices that can be activated by an
experimenter outside the subject's chamber. Sentences sent in to the subject by means of these
second projectors also go through, and are recorded
by, the computer; thus it will be possible, eventually, to program the computer to respond in Yerkish to some of the linguistic input from the subj ec t' s keyboard.
Sentence Analyser or Parser

Lexigrams are word-symbols in that they designate
concepts. They mostly correspond to English words
but, unlike English words, they are strictly univocal; ioe., each lexigram has one and only one
definition or meaning.
The rules governing the formation of sentences
in Yerkish are defined by a correlational grammar
derived from the correlational grammar of English
10

The grammatical analysis of input sentences is
carried out by a reduced and suitably adapted version of the Multistore Parser (von Glasersfeld and
Pisani, 1970) that constitutes the operational implementation of the correlational grammar of Yerkish. In the present system both a vocabulary of
250 lexigrams (half of which can be represented on
the keyboard at anyone time) and a grammar of maxCOMPUTERS and AUTOMATION for July. 1973

imally 46 connective functions and 46 conceptual
lexigram-classes are contained in approximately 4000
bytes of central core in a PDP-8E computer. Sentences up to a length of seven lexigrams are accepted and are fully parsed in about one second.
Every linguistic input is recorded in print-out and
various data are added. Date and time of input,
source of input (subject or experimenter), result
of sentence analysis (correctness, type of error)
are automatically recorded as well as the response,
if any, on the part of the computer.
According to the outcome of the parsing, the subject also receives acoustic feedback as to the grammaticality of the input sentence; a bell sound indicates correctness; a buzzer, errors (optional).
Automatic Computer Responses

Requests for various types of food and drink, for
a toy, for acoustic entertainment (primate vocalisations, different kinds of music), visual entertainment (sli~es or movies), and for opening the
window (view of outside world) are handled automatically by the computer. If the request is passed
as grammatically correct by the parser, the sentence
structure, in conjunction with the lexical items involved, triggers a coding device which then sends a
signal to the required dispenser or mechanism in the
experimental chamber.
If more than six lexigram keys are depressed before the period sign, the parser triggers the error
signal, records excess of length for the sentence,
and blanks the feedback projectors. Similarly, if
the period sign is activated and the parser has not
found a coherent syntactic structure incorporating
all the preceding lexigrams, the error signal is
triggered, the erroneous sequence is recorded, and
the feedback projectors are blanked.
Progress of the Subject

After several weeks of preliminary training in
discrimination tasks involving single Yerkish lexigrams (roughly equivalent to. the initial learning
of single words in the human infant), then single
lexigrams preceded by "PLEASE" and followed by "PERIOD," Lana was presented with fixed lexigram sequences representing simple requests such as "PLEASE
MACHINE GIVE M&M." Of;-M candies are still her
favourite item~). One press would activate the
whole sequence of keys and produce the reward. As
soon as she became proficient at obtaining the reward in this manner, the sequence of lexigrams was
split into two groups which now had to be depressed
one after the other. When this was learned, a further split was made, and so on, until all lexigram
keys had to be depressed singly and in the correct
sequential order before the reward was delivered.
This training wethod was used because it seemed to
be the best approximation to children's use of
"holophrases" (i.e., one word where an adult would
say a whole sentence). It turned out to be extremely efficient. When the first properly functionin~
keyboard panel of 25 individual lexigrams was presented to Lana, she had little difficulty in maintaining the correct lexigram sequence of her requests in spite of the fact that the physical position of individual lexigrams in the keyboard was
frequently changed. In other words, what she was
learning was manifestly not a specific motor pattern, but actually the temporal or sequential order
of the lexigrams.
This was further demonstrated by her ability correctly to finish sentences, the first two or three
COMPUTERS and AUTOMATION for July. 1973

lexigrams of which were activated by the experimenter. If, for example, the experimenter activated
the lexigrams "PLEASE TIM GIVE ... " in the first
three feedback projectors, Lana would complete the
request by depressing the keys corresponding to
" ... PIECE OF BANANA" or "JUICE" or "MS-M," and not
keys such as "MUSIC" or "WINDOW OPEN," which would
not have yielded a correct sentence (in Yerkish they
would require the beginning "PLEASE MACHINE MAKE
... "). This unexpectedly quick discrimination of
correct versus incorrect sentence completion has
prompted a special, fully controlled and statistically evaluated study which is now under way (Rumbaugh, Gill and von Glasersfeld, in preparation).
At the moment, about four months after her first
exposure to the system, Lana spontaneously makes a
variety of requests involving sentences of six lexigrams plus period, and she uses three activity lexigrams ("verbs"), each of which requires the choice
of a grammatical "object" that would be incorrect
for the other two activities. In all, she is now
using some 40 lexigrams.
On the whole, we have to admit that her progress
has been considerably faster than we expected. Just
now we are beginning to introduce her to "reading,"
i.e., to responding to Yerkish messages and instructions that are sent in to her by the experimenter.
Her first reactions to this new mode of communication are promising - so promising, indeed, that we
are beginning to wonder whether the technical work
involved in expanding the vocabulary, the keyboard,
and the automatic response mechanisms will be able
to keep up with the expansion of her skills.
In spite of her rapid progress, however, we have
not yet reached the point where we could gather evidence that a nonhuman primate is capable of interactive communication, i.e., capable of linguistically answering questions or participating in a conversation. All we can say at this stage is that, given
Lana's preliminary demonstration of aptitude, success in that direction now seems a good deal more
plausible than it did at the outset of the project.
References

Chomsky, Noam. Language and Mind. New York: Harcourt, Brace & World, 1968.
Gardner, B. T. and R. A. "Two-way Communication
with an Infant Chimpanzee." Behavior of Nonhuman Primates, Vol. 4, eds. Schrier and StolInitz. New York: Academic Press, 1971.
Ploog, D. and T. Melnechuk. "Are Apes Capable of
Language?" Neurosciences Research Bulletin,
Vol. I 9, IJo. 5, 1971.
Premack, David. "On the Assessment of Language Competence in the Chimpanzee." Behavior of Nonhuman
Primates, Vol. 4, eds. Schrier and Stollnitz.
New York: Academic Press, 1971.
Rumbaugh, Duane M., Ernst von Glasersfeld, Harold
Warner, P. P. Pisani, Timothy Gill, Josephine
Brown, and C. L. Bell. "A Computer-controlled
Language Training System for Investigating the
Language Skills of Young Apes." Antecedents of
Man and After: Comparative Behavioral Perspectives on Man. Vol. I: Primatology. The Hague,
Netherlands: Mouton, in press.
von Glasersfeld, Ernst. "The Correlational Approach
to Language." Pensiero e Linguaggio, Vol. 1, No.
4, InO.

von Glasersfeld, Ernst and P. P. Pisani. "The
Multistore Parser for Hierarchical Syntactic
Structures." Communications of the Association
for Computing Machinery, Vol. 13, No.2, 1970.

o
11

Computer Privacy and Security
Howard Campaigne
Lance J. Hoffman
Mathematics Department
Slippery Rock State College
Slippery Rock, Pa. 16057

"Several firms are now marketing as standard products bulk memories which could store a
one-page dossier on each of the 200 million citizens of the United States in ... about
225 square feet. In one system ... all of these dossiers could be available on-line, with
an access time of approximately 6 seconds. "

Computer systems have come a long way since the
first large scale electronic computer was developed
at the University of Pennsylvania in 1946. Today
computer systems can perform millions of operations
each second. They are used in process control, inventory management, information retrieval, war gaming, urban systems simulations, and many other diverse fields, in addition to relatively well-established areas such as payroll computation and account
billing.
Along with this increasing use of computers in
industry and government,a slow but steadily increasing public concern has developed over potential
threats to privacy posed by the advent of the computer. Fears of "1984" or "Big Brother" have developed, as well illustrated by the Newsweek of July
27, 1970. Its cover features a computer system
wearing an Uncle Sam hat and equipped with (as peripheral devices) a tapped telephone, videotape
camera, sound recording tapes, microphone, and other
listening or snooping devices. All of these are focused on one poor couple illustrated in a spotlight.
This magazine sitting on every drugstore and supermarket rack in the country focused the attention of
millions of Americans on privacy, data banks, and
the computer -- and its latent power for good or
evil.
The worry that some members of the public have
expressed about huge memory banks which contain personal dossiers is well founded, from the technological point of view. The technology for storing
enormous amounts of data in computer systems is
here. Several firms are now marketing as standard
products bulk memories which could store a one-page
dossier on each of the 200 million citizens of the
United States in a small area -- about 225 square
feet. In one system already delivered by Precision
Instruments, Inc., all of these dossiers could be
available on-line, with an access time of approximately 6 seconds.
A later .system, that of Ampex in Sunnyvale, California, has a l~ trillion bit wewory which uses
videotape recording techniques; it allows on-line
accessing of the equivalent of up to 32,000 reels
of tape or 350 IBM 3330 disks. This amount of memory is equivalent to over 30 double-spaced pages

12

per person for everybody in the United States.
erage access time is under 10 seconds.

Av-

IBM has delivered a special order system, the IBM
Mass Photodigital Storage System, which consisted
of trays of film chips. Each small 3~ mm. x 7 mm.
0-3/8" x 2-3/4") chip contained 4.72 million bi ts
of memory (read-only after once written). In each
cell was contained 151.04 x 10 6 bits and in an entire system one could have either 3.36 x lOll or
1.0 x 10 12 bits.
The average access time here was 10 seconds. One
could write as little as one file per frame (there
were thirty-two frames on a chip). There were
206,640 bits/frame and the bit size was 8.5 x 8.0
microns.
The Grumman MASSTAPE system can store 16 billion
bytes in 16 square feet. Up to eight of these units
may be combined to provide a trillion bit (128 billion byte) system at a cost of about .0001 cents per
bit (= $1 million).
As we have seen, the technology in memories is
here. How extensive, then are the dossiers currently kept in computer banks? What data is kept in
them and who has access to this data?
If we take a look at the panorama of data banks
which exist today, we find that the United States
has become a records-oriented society. Records are
kept on our education, employment, credit, health,
and in government files such as taxation and law enforcement. During the past two decades, as most
government agencies and large private organizations
have been computerizing their large-scale files, the
American public has become concerned that because of
the computer's enormous capacities far more personal
data might be assembled about individuals than it
had been feasible to collect before, and that at the
same time there might be a decrease in an individual's ability to see and challenge records on himself .
The fact is that computerized data banks
contain sensitive information on people are
nearly as advanced as one might think after
some of the literature warning of potential

which
not
reading
misuse

COMPUTERS and AUTOMATION for July. 1973

of these systems. The VISIon of a general-purpose
total management information system providing key
executives the information they need on a cathode
ray tube at the touch of a button -- the image so
popular in much of the glossy literature of the 60 i s
is still just that -- a vision. In practice, it
is nowhere to be found in 1973.
More important, the most sensi tive information -information which is text (character strings) rather
than numerical values and which is therefore more
likely to be subjective -- is usually not going into
computer systems. With the technology that has been
available to date, it has simply been too expensive
to store large amounts of text in most computer systems. For a combination of technological and organizational reasons, central databank developments are
far from being as advanced as many public commentaries have assumed. Organizations have so far
failed to achieve the "total" consolidation of their
information about individuals which raised civil
liberties alarms when such goals were announced in
the 1960's by various government agencies or private
organizations.
Further, in computerizing their records on individuals, organizations have generally carried over
the same policies on data collection and sharing
that law and adminis trati ve tradi tion's in each field
had set in the pre-computer era. Where new law or
practices have evolved to protect individual liberties over the past decade, organizations with computerized systems have followed such new policies
as fully as those that still use manual files and
procedures. Even the most highly computerized organizations continue to rely heavily on manual
record keeping and retain in their paper files the
most sensitive personal information they possess.
We should not infer, however, that organizational
and technical limitations will hold down the amount
of information in computer databanks forever. Precisely because of the technological advances in memories mentioned earlier, it is becoming economically
feasible to store more and more data in relatively
fast on-line memories. As the cost of these memories is driven down, more and more textual and subjective information is likely to be stored in computer systems.
Some areas where data banks with sensItIve information on people already exist are law enforcement
(e.g., the FBI's National Crime Information Center
Computerized Criminal History file), retail credit
bureaus (e.g., TRW-Credit Data Corporation's completely computerized credit history data bank) and
state welfare system data banks.
These and other computer systems all reflect the
age-old problem of resolving the conflict between
privacy on the one hand and data acquisition for use
in better and more rational decision-making on the
other hand. One has to make trade-offs. If society
desires better medical systems, speedy and accurate
airlines reservation systems, less policemen shot
when they stop cars, or faster distribution of welfare checks (just to give a few examples), it will
have to give some "private" data to the computer
systems which provide these advantages, and thus increase the risk of unauthorized disclosure in order
to gain the benefits. Note that sometimes the risk
is decreased in computer systems as opposed to manual systems. Data is often better protected in computer systems than in manual file folders presided
over by flatter-prone clerks working for low wages
in large offices.

COMPUTERS and AUTOMATION for July, 1973

As reported by the National Academy of Sciences
Proj ec t on Compu ter Databanks in 1 a te 1972, "the
real issue of databanks and civil liberty facing the
nation today is not that revolutionary new capacities for data surveillance have come into being as
a result of computerization. The real issue is that
computers arrived to augment the power of organizations just when the United States entered a period
of fundamental debate over social policies and organizational practices and when the traditional authority of government institutions and private organizations has become the object of wide-spread
dissent. Important segments of the population have
challenged the goals of major organizations that use
personal records to control the rights, benefits,
and opportunities of Americans. There is also debate over the criteria that are used to make such
judgements (religious, racial, political, cultural,
sexual, educational, etc.) and over the procedures
by which the decisions are reached, especially those
that involve secret proceedings and prevent individuals from having access to their own records. Computers are making the record-keeping of many organizations more efficient precisely at the moment when
trust in many large organizations is low and when
major segments of the American populations are calling for changes in values that underlie various
social programs, for new definitions of personal
rights, and for organizational authorities to make
their decision-making procedures more open to public
scrutiny and to the review of specific individuals
involved."l
The report identified six areas of priority for
public action:
1. Development of laws to give the individual a
a right of access and challenge to almost every
file in which records about him are kept by government agencies.
2. Development of explicit laws or rules which
balance confidentiality and data-sharing in many
sensitive private record systems which today do
not have such rules. These include bank records,
and travel and charge card records.
3. Limiting the collection of personal information and its use in such areas as arrest-only
records in licensing and employment decisions;
and selling of name and address lists collected
by government agencies to mailing list sellers
(unless the individual consents).
4. Increased work by the computer industry to
make technological safeguards "available and
workable products".
5. Reconsideration by Congress and the Executive
Branch of the "current permissive policies toward
use of the social security number". The report
contends that a minimum level of trust must be
maintained between American citizens and their
government. "Under these conditions, adopting
the social security number as a national identifier or letting its use spread unchecked cannot
help but contribute to public distrust of government. "
6. Experimentation with special information-trust
agencies. For example, the handling of national
summary criminal histories (rap sheets) might be
taken away from the FBI and given to an independent national agency under control of a board
which would have public representatives as well
as law enforcement officials on it.

13

In its closing paragraphs, the report sums up the
data banks and civil liberties problem as follows:
"If our,empirical findings showed anything,
they indicate that man is still in charge of the
machines. What is collected, for what purposes,
with whom information is shared, and what opportunities individuals have to see and contest records are all matters of policy choice, not technological determinism. Man cannot escape his
social or moral responsibilities by murmuring
feebly that 'the Machine made me do it'."
Safeguards for Security

To turn now to security measures in computer systems, one must first realize that most safeguards
for privacy are not delivered as standard, off-theshelf items when computer systems are bought or
rented today; they do not come with the hardware or
with the software. Usually little or no mention of
privacy problems is made when the computer system is
contracted for. If security safeguards are supplied,
they are generally furnished as special optional
items, tailored to each individual installation with
commensurate cost to that installation. Since concern about the problems of personal privacy in computer systems is now just beginning to surface, manufacturers are just now starting to seriously address these problems. But today there is not one
standard system - that is, one you can buy "off
the shelf" with no optional hardware or software which contains adequate privacy safeguards for personal data.
Probably the most important controls of all are
administrative, not technical. Administrative controls determine what gets collected in the first
place and to a large extent who it is disseminated
to once it is collected. Even then, one should realize that no system will even work exactly as designed; we live in the age of the Xerox machine, and
in the age of the information buddy system. This
buddy system functions in any information system,
and there will probably always be a few unauthorized
leaks. Interestingly enough, computer systems are
very possibly more secure than manual systems, since
with the advent of the computer, people must be instructed in how to use the computer system. No
longer is it the case that anyone who can read can
open the file cabinet and scan his neighbor's folder.
Another important nontechnological control method
is physical security. This can be very important
and is often overlooked. One may be using all the
passwords and other technological safeguards, but
without physical security at the computer center and
at the files, one's efforts will be for naught.
There is quite a bit of literature around now on
physical security, one of the best short works being
IBM's booklet "The Considerations of Physical Security in the Computer Environment".2
Technical Methods for Security

Assuming that physical and administrative safeguards have been taken, one can now turn to technical methods for security in computer systems.
Authentication and Identification

Current software identification techniques are
generally based on something the user knows, something he has, or something he is. Comparisons are
made between the proper "something" and what the
computer expects, and a decision to permit or deny
access is made accordingly.

14

In the category of "something he knows", there
are several techniques. Simple passwords offer protection against what might be called "casual" infiltration. They are used to protect files, records,
data fields within records, or groups of these items.
Once the computer verifies that a proper password
can be associated with a given user, that user can
obtain all the information authorized to that password.
While the simple password scheme is easy to use
and inexpensive to implement, it requires the authorized user to remember the password and it can be
compromised by wiretapping, electromagnetic pickup,
or (more simply) an interloper discovering the password written down somewhere, such as on an office
calendar, a scrap of paper tossed in the wastebasket, or on discarded console output. There is also
an inverse relationship between the length of time a
password is in use and the degree of security it
maintains. For this reason, passwords should be
changed aperiodically to thwart unauthorized users.
Passwords should not be printed (echoed) at consoles; overprinting them there is not always effective. Passwords should also undergo some minimal
encryption before being sotred in computer systems.
There are a number of variations on password
schemes. One of these which eliminates some of the
disadvantages named above is the following. The user
remembers a password as before. When signing on, he
is requested by the computer to type certain characters of the computer's choosing. In this way the
possibili ty of compromi se by ribbon reading or wastebasket searching is greatly reduced. However, all
the other disadvantages of the passwords remain.
One-time passwords can also be used. These protect against stolen password utilization because
once a password is used it is no longer valid.
Further along the spectrum, many disadvantages of
passwords are not present if the computer system requires successful execution of an algorithm for the
authentication process. This is often called a
handshaking procedure. Questions asked by the computer can be personal information ordinarily unknown
to anybody but the authorized user (such as his
brand of hair tonic). Alternatively, the user can
be asked to apply a secret function to a pseudo-random number generated by the computer. The function
is never transmitted in the clear over communication
lines or typed on a console.
Electronic combination locks are also available
(from, for example, Datalock 3 ) where a touch-tonelike keyboard is used. The user must press the
right combination to physically enter the area of
the computer terminal.
In the category of "something he has", the user's
photograph and signature can be attached to a credit
card-like device. This prevents illegal users from
easily passing a guard and getting to computers or
remote consoles. These credit card-like devices
come complete with threat-monitoring printer units.
For example, one access control system is a locking
and identification unit which is like an ordinary
door lock and key except that the key is a binary
coded printed circuit. A central printer can record the date, time, key identification number, and
door number of each entry.
There are also some devices that allow access to
a user based on "something he is". In this category
we have the Identimat,4 which identifies individuals

COMPUTERS and AUTOMATION for July, 1973

by measuring hand geometry. This system claims 99.5
percent accuracy but the cost per station is relatively high. Experimental work has also been done
on voice prints, though the results in this area are
to date far from conclusive.
Another authentication method is the dial-up and
call-back method. When data is desired by a user
who is connected to the computer via telephone line
A, a message is sent to the user asking him to provide the password of that data to the computer
operator by voice over a different telephone line B.
Alternatively, the operator calls the user back over
phone line B to make sure the user is authorized and
at the right place.
In addition to the authentication methods which
are used to control access to files, or to the computer system itself, other methods are necessary to
handle shared data bases and files which contain information needing different degrees of protection.
Compartmentalization Methods

Various technical methods for compartmentalizing
computer memories are in use. By compartmentalizing, we mean those methods used to subdivide the
memory into parts with different access privileges.
We use the inelegant word "compartmentalizing" because words in more common use (e.g. partitioning or
segmenting) are too closely linked to specific hardware. There are two types of compartmentalizing
techniques; those which are necessary for the proper
operation of the computer system (especially a timesharing system) and those which are used to enhance
information security in a shared system.
Hardware. In any mUltiprogramming, multiprocessing or time-sharing system each user must be prevented from disturbing the program or data of the
others. Furthermore, a user should not be allowed
to interfere with operations by improper or untimely
input or output or halt commands. This latter restriction is generally obtained by denying to the
user certain instructions which may be executed only
by privileged programs such as the operating system.
The former capability generally is provided by memory protection schemes such as relocation and bounds
registers, segmentation, paging and memory keys.
The techniques mentioned in the preceding paragraph protect contiguous portions of (real or virtual) computer memory from alteration by an errant
or unauthorized program. These hardware countermeasures do not, however, provide protection of a
user file in auxiliary storage from unauthorized access. To this end, software schemes have augmented
the hardware methods.
Software. Only a few installations are using the
more sophisticated of the methods described below.
Computer users are pleased (even astonished) when
their programs work, and most -- if they use any security protection at all -- use simple passwords to
guard their files. In many cases this level of protection is all that is justifiable. More detailed
descriptions of some of the methods mentioned below
appear in an anthology on computer security.s
A system involving concentric "rings" for protection developed at MIT has recently been announced by
Honeywell as an upcoming product. It provides flexible but controllable access by a number of different users to shared data and procedures. On the
negative side, as implemented, it requires hardware
segmentation; and if a file has many different data
fields with many different levels of access, the
COMPUTERS and AUTOMATION for July. 1973

swap times necessary to access each datum in its own
segment cause prohibitive overhead. It also imposes
a hierarchy on all information in the data base;
this is not desirable in every instance.
Software compartmentalization below the file
level is not in general commercial use. This is
probably due in varying degrees to the obscurity of
the methods, the lack of reliable information on
overhead costs, and until recently, manufacturer and
user indifference twoard data security.
Privacy Transformations

Privacy transformations are reversible encodings
used to conceal information. They are useful for
protecting against interception on communication
lines, neutralizing of electromagnetic radiation
from terminals or computers, or wrongful access to
data in removable files. This last threat is probably the most significant.
Simple privacy transformations are one of the
most effective methods of preventing disclosure of
data. Even if an infiltrator somehow obtains the
data he must still decrypt it; often he does not
have the time or resources to do this, and even if
he does, he must decrypt a mass of uninteresting
data just to find what he wants. It would usually
be much simpler to bribe a legitimate user.
Example of a very simple (and fairly insecure)
privacy transformation:
Cleartext:

PLANES ATTACK AT 0500.

Key:

3141631416314163141631

Ciphertext (using 1T as the "key" and "addition"
a s the method):
SMEOKV. EUZDDO.GW.4663A
Privacy transformations can be effected by either
hardware or software. Scrambling devices which use
group theory, pseudorandom number techniques and
other methods are now available commercially from,
for example, Ground Data in Ft. Lauderdale, Fla. and
Technical Communications Corp. in Lexington, Mass.
These boast hardware at the terminal and software or
hardware at the computer. More detailed information
on privacy transformations can be found in Kahn's
excellent work The Codebreakers 6 and in Hoffman's
anthology.
Threat Monitoring

Software, when used in conjunction with administrative rules and compartmentalization methods which
protect the operating system from unauthorized
changes, can detect attempted or actual security violations, and either provide an immediate response
(e.g. job cancellation or tracing procedures) or
provide an ex post facto analysis. It can also record all attempts or all rejected attempts to obtain
certain sensitive information, unusual activity of a
given file, attempts to write into protected files,
or excessively long periods of use. In fact the
last is perhaps the only known method to cope with
the fact that, given enough work and some prior information, it is possible to obtain a specific "dossier" from a "statistical" data bank -- the "statistical inference problem".
The use of such software is called "threat monitoring". The log kept is also called an "audit log"
or an "audit trail". The monitoring or auditing

15

provides a method for recording the identification
of each request and the name of each requestor. The
audit trail can be examined periodically by appropriate authority to determine if unauthorized access
has occured and then suitable steps can be taken.
Threat monitoring has a beneficial side effect;
it often will help improve the efficiency of the
system by recording widespread use of particular facilities. These facilities can then be altered to
eliminate bottlenecks. Also, if some security restriction is unduly interfering with operations,
threat monitoring should help pinpoint the offending restriction.
Costs of Countermeasures

General. There is almost no information available on the cost or overhead of security techniques
for computer systemso Fixed costs of security controls are very dependent on the vicissitudes of the
particular installation. These include the capital
cost, the one-time outlay to get into operation, and
the expenses of buying, building, or programming a
system. These also include administrative costs
such as costs of guarding doors, having telephones
elsewhere, and perhaps even having some people devoting their time to penetrating the system.
Many of the costs of security are variable, even
at a given installation. These costs depend on the
frequency with which records are used, since this
determines how often some safeguards are invoked.
In addition to these fixed and variable costs,
there are other costs even harder to quantify; the
development cost, the expense of experimenting and
rejecting approaches before finding an acceptable
one; the social cost of being secretive; the social
cost of not being secretive; the cost of doing some
functions manually because adequate security controls are not available in the computer system; and
others. It is impossible to evaluate or even enumerate these latter costs.
Administrative Rules

Administrative rules do not come free; they take
effort on the part of the administrators. Also, restrictions of access to tapes, documents, or other
items must be enforced by employees who have other
tasks to perform as well.
Such precautions as logs and artificial data for
demonstrations cost money to generate and maintain,
as well as requiring constant alertnesso Accounting
for sensitive punched cards, tapes, printed forms,
carbon paper, and ribbons is not very expensive, but
is not free either, for someone must be responsible
that pains continue to be taken that this chore be
done and done thoroughly.

been "soaked" with ultrasonic traps and other measures at a cost of about $12,000. A recent Wall
Street Journal account states that Data Processing
Security, Inc. of Hinsdale, Illinois, "will put a
team of consultants to work running through a 172point (security) checklist and preparing a survey
report for $3,000 to $5,000."
One can use double-door "buffer" systems with
electric locks, magnet sensors and closed-circuit
television to control access to computer centers. In
contrast, locks and keys can also be obtained for
under $100.
Partitioning for security may mean somewhat inefficient use of space, another cost. Identification
badges cost very little, but they do require someone
to supervise, monitor, and maintain them, a halftime job for even a small installation. Very small
installations can rely on personal identification.
Authentication Methods

Hardware in terminals which identifies the terminal or the user to the system is still an optional extra in most cases.
The ordinary password is an inexpensive method,
well tested and perfected by much use. It can be
implemented on existing systems without special
hardwareo The use of one-time passwords is not
nearly so well known, is more secure, and involves
little incremental software cost over the simple
password, but it is much harder on the user, who
must know either by memory or by algorithm the current password.
A handshaking algorithm is a much more elaborate
procedure than any other involving the user. It
takes more of his time, more communication line
time, and more programming effort. It is, of course,
more effective and worth what it costs to those who
need it. Sirice the complexity of handshaking algorithm is adjustable to the requirements of any given
system, the procedures can meet a large variety of
needs, and offer much promise for the future. But
for the same reasons, they can be arbitrarily expensive.
The use of simultaneous software keys from different terminals, which has been suggested as a very
secure procedure, throws a quite demanding requirement on the coordination of the people involved. The
cost of equipment and of software to do this is relatively low. But the necessity for the people involved to be at certain places at certain times will
limit this procedure to very unusual situations.
The dial-up and call back method is relatively
cheap for the protection it provides.
Compartmentalization Methods

Physical Security

The most elementary precaution is to control physical access to the system and its files. This means
either arranging partitions and doors so as to limit
the approaches and arranging for locks and keys, or
else having personnel in the area night and day. The
latter is very expensive if the workload would otherwise not demand it.
The costs of elaborate physical security measures
vary widely, depending on the particular installation. Newspaper accounts have stated that the William J. Burns International Detective Agency data
processing center in Briarcliff Manor, New York, has
16

Hardware compartmentalization of memory is necessary for the proper operation of any multiprogramming system; therefore that facility is free, so to
speak.
The use of individual field authorizations is not
widespread, and costs are ~ot completely known. The
incremental costs of methods which allow individual
field authorizations are unknown, although they are
clearly at least as expensive as those methods which
do not allow them. Even if they cannot be made
cheaper than the other methods, they still offer
much promise because of their flexibility and comprehensiveness.
COMPUTERS and AUTOMATION for July. 1973

Privacy Transformations

Methods of encrypting are such that a good deal
of security, measured by the amount of work required
to break a cipher, can be had with only a modest increase in expense. Some recent work has shown that
simple cryptographic encoding can be done at rates
of about 50,000 bytes/second on medium-sized computers.
Products for encrypting transmission over vulnerable communication lines are currently available.
Ground Data Corp. of Ft. Lauderdale, for example,
was in 1972 selling such a system with immediate delivery for $975; rental of the same unit was $35 per
month.
Privacy transformations lose much of their utility if an entire file must be decrypted and re-encrypted using a different transformation many times.
This can happen if an installation has a large encrypted file and changes encryption frequently in
order to maintain a high level of security. The computer time required by this can be quite considerable. Frequent passing of a large and possibly multi-volume file, the only effect of which is to
change the encipherment, is nearly always prohibitively expensive. For this reason many experts feel
that cryptography will not generally be used for encrypting entire data sets but will be used for transmissions over communication lines, where encryption
and decryption are done only once. Frequent re-encryption of an entire file can be avoided by providing adequate physical security of the place where
the file is kept and of the transmission channels,
where applicable. Here, as usual, resources saved
on one safeguard (privacy transformations) can be
used to strengthen another safeguard (physical security) and thus achieve better results.

NUMBLES
Neil Macdonald
Assistant Editor
Computers and Automation
A "numble" is an arithmetical problem in which: digits
have been replaced by capital letters; and there are two
messages, one which can be read right away and a second
one in the digit cipher. The problem is to solve for the
digits.
Each capital letter in the arithmetical problem stands for
just one digit 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 transla ted (using the same key)
into letters so that it may be read; but the spelling uses
puns or is otherwise irregular, to discourage cryptanalytic
methods of deciphering.
We invite our readers to send us solutions, together with
human programs or computer programs which will produce
the solutions. This month's Numble was contributed by:
Andrew M. Langer
Newton High School
Newton, Mass.

NUMBLE 737
TOO

1= W

+ S 0 F T

8 = F

Q Z U D

L=T

Threat Monitoring

The costs of threat monitoring are not known accurately now, but they are principally the cost of
deciding what information should be kept in logs,
the operating cost of examining those logs, and the
maintenance of the ~oftware. Sufficient magnetic
tapes or other storage on which to record the audit
trails will be required, and additional system overhead will be caused by the recording of accesses and
refused admission.

+ WI

L L

D Z E T E

+

8 E

D 0 U Q L

4509 9291

Summary

Solution to Numble 736

In summation, security of private or sensitive
information in computer systems involves legal, administrative, physical, and technological controls.
Often one can arrive at a desired level of security
by trading off safeguards in one area (such as technological) to increase them in another (such as physical). In the final analysis, this decision must
be made by each installation based on its own needs.
References

10 Westin, A.F. and M.A. Baker, Databanks in a Free
Society, Quadrangle Books, New York, 1972.
2. IBM, "The Considerations of Physical Security in
the Computer Environment", Form G520-2700-0.
3. Datalock, Sacramento, Cao
4. Identimation Corp., Northvale, NoJ.
5. Hoffman, L.J., Privacy and Security in Computer
Systems, Wiley, 1973
6. Kahn, David, The Codebreakers, The Macmillan Co.,
New York, 1968
[]

COMPUTERS and AUTOMATION for July. 1973

In Numble 736 in the June issue, the digits 0 through
9 are represented by letters as follows:
U, V, D = 0

C=5

R= 1

Y=6

H=2

E=7

A=3
T=4

M, N

=8

S=9

The message is: The years teach much that the days never
teach.
Our thanks to the following individuals for submitting
their solutions - to Numble 735: Edward A. Bruno, N.
Bergen, N.J.; John F. Gugel, Arlington, Va.; Felicitas
Reich, Yonkers, N.Y.; Abraham Schwartz, Jamaica, N.Y.

17

Computer Programming Using Natural Language
-

Part 2

Edmund C. Berkeley, Andy Langer, and Casper Otten
Associates, Frontiers Group
815 Washington St.
Newtonville, Mass. 02160

'The name 'GENIE' comes from the famous genie or jinni of the Arabian Nights tales, who,
whenever summoned out of his bottle by his master, the possessor of the bottle, replied,
'1 hear and obey,' without requiring translation of the commands into any other language."

Outline

1. A Computer Program to Understand Manager's Instructions to a Clerk
2. Calculation Layout Language
3. First Illustrative Problem, Mean and Standard
Deviation
4. Second Illustrative Problem, Loans and Interest
5. Third Illustrative Problem, Advertising Effectiveness
6. Future Dialog with a Businessman
1. A Computer Program to Understand
a Manager's Instructions to a Clerk

This article is a continuation of the article
"Computer Programming Using Natural Language" which
was published in the June issue of Computers and
Automation. But the present article has been written independently; though access to the prior article is helpful, it is not necessary.
The subject we are dealing with here is a part of
the subject of "computer programming using natural
language," and consists of a certain task. The task
that we have set ourselves is called "Task F"; and
it may be described as follows:
Produce a definite, efficient, and rather
small computer program (let's name it GENIE)
which will take in ordinary natural language
specifying anyone of many kinds of computer
application problems, and which will put out
automatically a computer program in a specified, regular programming language which will
correctly handle any sample problem of that
kind.
The name "GENIE" comes from the famous genie or
jinni of the Arabian Nights tales, who, whenever
summoned out of his bottle by his master, the possessor of the bot tIe, replied, "I hear and obey,"
without requiring translation of the commands into
any other language!

18

The input that we give to the computer program
GENIE so far consists of:
(1) a manager's instructions to a clerk in
ordinary natural language,
(2) one or more worked numerical examples, and
(3) a calculation layout form.
The final output that we obtain from the computer
program GENIE is a computer program in a regular
programming language, such as assembly language or
BASIC. This output program-has the property that
it will handle correctly all the kinds of calculation specified by the manager's instructions, the
worked numerical examples, and the layout form.
GENIE takes in a passage of ordinary natural language written with a very wide degree of freedom,
and produces a computer program which will handle
all the applications of a given kind described or
envisioned by that passage. Thus the job of conforming to the rigorous requirements of writing a
successful program in a regular programming language
like BASIC or FORTRAN is largely lifted from the inaccurate human being, and delegated to the accurate
computer under control of GENIE.
It must be emphasized that we are at the beginning of this project, with only a few examples successfully handled so far; but we see no theoretical
barriers. And the next version of GENIE should be
considerably more powerful than the current one.
In the prior article we presented one illustrative application problem in full. That problem was
the calculation of the mean and standard deviation
of a frequency distribution.
In this article we present
application problems, showing
and what comes out of GENIE.
simple problems in the fields

two more illustrative
what goes into GENIE
These two problems are
of

-calculation of interest on loans and amount
to be repaid;

COMPUTERS and AUTOMATION for July. 1973

j
I

Statement to the press by Edmund C. Berkeley, Andy Langer, and
Casper Otten on June 11, 1973.

We believe we have accomplished a breakthrough
in programming computers.
We have taken ordinary instructions in natural
language such as a manager would give a clerk in
order to do a calculation, a worked example of the
calculation, and the layout of calculation. Then,
using a computer program which we call GENIE, we
have produced an accurate computer program which
will cause the computer to do all of that kind of
calculation.
The work leading to this new result is an offshoot of research on computer-assisted documentation of Navy computer programs done under a research contract from the Office of Naval Research.
We are only at the beginning of a long road -getting computers to understand human beings
speaking in ordinary natural language, But lots
of people will no longer need to learn BASIC or
FORTRAN or some other special language in order
to tell a computer what they want it to do,
What we have done is like a proof of possibility in mathematics -- the exhibition of a result
that has certain properties.
The reason for our breakthrough is that we have
been using a number of principles all together,
mathematical, linguistic, grammatical, logical,
practical, common sense, and so on. The solution
to ordinary natural language programming seems to
be easier when you use enough principles together,
If an ordinary clerk can understand an ordinary
manager, then a computer program ought to be able
to do as well. We have demonstrated that in several cases.

really very close to what may be called Calculation
Layout Language. This is a language which is used
extensively in the business world in forms and on
calculation sheets, to guide clerks and to tell the
relations of the items on a form, and the relations
of the column headings on a calculation sheet, This
language uses words or numbers or symbols to express
concisely (but not always completely or uniquely)
the calculating rules being used.
An example of Calculation Layout Language is
"(4) = (1) x (3)," where the "x" means "times", (In
Level Two Language, the corresponding statement is
(4) = (1) ':' (3), where the asterisk means "times";
the change is due to the necessity of distinguishing
"x" as a letter of the alphabet and "x" as an ideographic symbol for "multiplied by".)
The statement "(4) = (1) ':' (3)" asserts either
one of two meanings depending on context. The first
meaning is "Item 4 equals Item 1 mul tiplied by Item
3". The second meaning is "Each figure in Column
(4) is equal to the corresponding figure in Column
(1) multiplied by the corresponding figure in Column (3)". In business practices, these two meanings
are distinguished only when the context so requires;
ordinarily both meanings are considered "evident,"
"understood," "obvious," in making business calculations on layout sheets.
A characteristic feature of Calculation Layout
Language (and Level Two Language) is the designation
of items or columns by numbers written in parentheses. Occasionally Calculation Layout Language contains some abbreviations such as FICA meaning "Federal Insurance Contributions Act deductions," i.e.,
the deduction from pay under the Social Security
system, At this time, GENIE is happy with any number in parentheses used as a name for a variable;
but at this time GENIE would have to be told in a
specific manner (it could not guess as an experi-

Computer Programming Using Natural Language
- Part 1, June, 1973, pp 9-15, ...

-calculation of the effectiveness of advertis ing,
We also discuss in a hypothetical dialog advantages and limitations of GENIE.
The concept of GENIE can, of course, be expressed
in many versions. In the future GENIE surely will
be programmed more efficiently than the versions we
have worked out so far. But in the meantime the
three illustrative examples constitute a demonstration, a proof, of the possibility of using ordinary
natural language to program a computer, and a refutation to those who say such a thing is not possible.

Table of Contents
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.

2. Calculation Layout Language

II,

GENIE has two outputs, deliberately. The first
output is something like a set of formulas, describing what is to be done. We as human programmers regularly review this output, because we can
understand it; we look to see if it makes sense and
expresses essentially what is to be done. The second output is a specific computer program stated in
a specified programming language which precisely
performs the calculations needed. The only quick
way to check this output is to try it on the computer with sample data and see if it produces the
right answers,

12.
13,
14.
15.
16.
17.
18,

The first output is expressed in a language that
we call Level Two Language. Level Two Language is

COMPUTERS and AUTOMATION for Ju Iy. 1973

Automatic Programming Using Natural Language
Task F
Frontiers Group
The Starting Point, and GENIE
A Specific Example
Strategy
The Principle of Appropriate Neglect
The Principle of the Preferred Synonym
The Principle of Relevant Context
The Principle of a Series of Separate Statements
The Principle of Statements of Reasonable
Length
The Principle of a Finite Vocabulary
The Principle of Gradual Development
The Principle of Questioning by the Clerk
Other Principles
Level One Language and Level Two Language
Cri ticisms
The Future

Table: Probably Acceptable Vocabulary
Probably Unacceptable Vocabulary
(At This Time)
General Business Applications to Which the
Program GENIE Might Apply First

19

Table 1 of Article
Calculating Interest and Amount to be Repaid Manager's Instructions to a Clerk In put to G ENIE
1. To Fill in the Register (Form 2): Copy from
Mr. Richard Roe's "Notice of Loan Granted" (Form 1):
Loan No., 105; Register No. of Borrower, 777; Principal, $1750.00; Rate, 7~; Date Loaned, Aug. 1,
1972; Date Due, May 15, 1973.
2. To Fill in the Worksheet: Look up date of
loan, Aug. 1, 1972, in Column 1 of Table 1 (see below), and find the number of half months, 14, and
put that in Col Ulan 1 of the Worksheet.
3. Look up Date Due, May 15, 1973, in Table 1,
and find the number of half months, 33, and put that
in Column 2 of the Worksheet.
4. Next, fill in Column 3, which equals Column 2
less Column 1. The result is 19.
5. Then, divide Column 3 by 24 to get fraction of
a year, .7917, and put that in Column 4.
6. Take the rate of interest, 7YiYo, in Column 4 of
the Register, express it as a decimal to three decimal places, .075, and put that in Column 5 of the
Worksheet.
7. Take Column 4 and 5 and multiply them and put
the result, .05938, in Column 6.
8. Take the principal amount in Column 3 of the
Register, and multiply it by Column 6 of the Worksheet, finding the actual amount of interest, 103.92,
and put that in Column 7 of the Worksheet.
9. To Fill in the Due Payments Record: Column 1,
Loan No., is the same as Column 1 of the Register,
105; Column 2, Number of Borrower, is the same as
Column 2 of the Register, 777. Column 3, Date Due,
is the same as Column 6 of the Register, May 15,
1973.
10. Column 4, Amount of Interest, is the same as
Column 7 of the Worksheet, 103.92.
11. To compute Column 5, Total Repayment, find the
sum of Column 4 of the Due Payments Record and Column 3 of the Register, which results in 1853.92.

enced clerk could) that a word like "FICA,fl not
found in the "probably acceptable vocabulary," was
being used for the name of a variable.
Clearly, if a calculation to be performed can be
expressed in Calculation Layout Language instead of
(or in addition to) ordinary English, we may be able
in some cases to dispense with interpreting the
"manager's instructions to a clerk". The one or
more worked examples, and the calculation layout,
however, will probably still be needed.
3. First Illustrative Problem,
Mean and Standard Deviation
The first illustrative problem, which was explained in the prior article, was the calculation
of the mean and the standard deviation of a frequency distribution. We showed a sample of manager's instructions to a clerk, a fully worked example, and the calculation layout (as expressed in
20

Table 1 of
Manager's Instructions for
Second Problem
Half Months Corresponding to Calendar Date
0)
Date in
197L

Jan.
Jan.
Feb.
Feb.
Mar.
Mar.

(2)

No. of
Half Months

1
15
1
15

o
1
2

3

1

4

15

5

Apr. 1
Apr. 15
May 1
May 15
June 1
June 15

6

July 1
July 15
Aug. 1
Aug. 15
Sept. 1
Sept. 15
Oct.
Oct.
Nov.
Nov.
Dec.
Dec.

1
15
1
15
1

15

7

8
9
10
11

12
13
14
15
16

17
18

19
20
21
22
23

( 3)

Date in
1973

(4)
No. of
Half Months

1
15
15
15
1
15

24
25
26
27
28
29

Apr. 1
Apr. 15
May 1
May 15
June 1
June 15

30
31
32

July 1
July 15
Aug. 1
Aug. 15
Sept. 1
Sept. 15

36
37
38
39
40
41

Oct.
Oc t.
Nov.
Nov.
Dec.
Dec.

42
43
44
45
46
47

Jan.
Jan.
Feb.
Feb.
Mar.
Mar.

1
15
1
15
1
15

33

34
35

the example). This input going into a version of
GENIE produced two outputs. The first output expressed the calculating rule in Level Two Language
(Calculation Layout Language). The second output
was a computer program in BASIC which computes the
mean and standard deviation of a frequency distribution.
4. Second Illustrative Problem,
Loans and Interest
For a second illustrative problem, we take a simple business situation. The amount of interest on
a loan; arid the t~tal amount d~e to be repaid; are
to be calculated, under the following assumed conditions.
Richard Roe grants loans for periods less than a
year on various kinds of security at simple interest
rates varying from 5% to 10%, to various parties.
(See Tables 1 and 2.) He keeps a Register (Form 2)
which shows: loan number; identification number of
borrower; amount of loan that he grants; a simple
interest rate per annum that he chooses; the date
of the loan; and the date due for repayment. Roe
always makes loans on the 1st or the 15th of a
month, and always makes loans repayable on the 1st
or the 15th of a month. His assistant does the calculations systematically on a Worksheet (Form 3).
The date due, the amount of interest due, and the
total repayment are posted in a Due Payments Record
(Form 4). The instructions which guide Smith are
shown in Table 1. A sample "notice of loan granted"
(Form 1), and extracts from the Register, the Worksheet, and the Dues Payment Record are shown in
Table 2, displaying worked examples. The contents
of Table 1 and of Table 2 are the input to GENIE.
COMPUTERS and AUTOMATION for July. 1973

.'
,-

items or columns in a sequence. The other change
is to convert the dates and months into straight
numerical data. Both of these editing changes are
shown in full in Table 3. With a later version of
GENIE it will not be necessary for a human being to
make these editing changes, since the computer programming to accomplish them is minor. But for the
present it takes time to make the changes in the
program instead of by editing; and that time is diverted from direct work towards the main goal, which
is understanding ordinary natural language that expresses instructions. And so for the present, this
minor editing work is left on the shoulders of the
human being.

Table 2 of Article
Calculating Interest and Amount to be Repaid Worked Examples Input to GENIE

FORM 1: NOTICE OF LOAN GRANTED
Loan No.
Borrower No.
Principal Amount
Rate of Interest
Date Loaned
Date Due

1.
2.
3.
4.
5.
6.

105
777
$1750.00
7Y§o

Aug. 1, 1972
May 15, 1973

FORM 2: REGISTER OF LOANS BY RICHARD ROE
0)

(2)

(3)

(4)

( 5)

( 6)

Loan

Borrower
No.

PrinciQa1

Rate

Date
Loaned

Date
Due

101

777

$1750.00

7~

2/15/72

8/1/72

102

658

3600.00

8~

3/1/72

7/1/72

103

721

600.00

9%

3/1/72

11/15/72

104

746

8000.00

5~

3/1/72

12/15/72

105

777

1750.00

7~

8/1/72

5/15/73

~

FORM 3: WORKSHEET FOR CALCULATING INTEREST
(1)

( 2)
(3)
(4)
(5)
( 6)
(7)

.,.
I~

Half Months of Date Loaned
Half Months of Due Date
Half Months of Period
Fraction of Year
Annual Interest Rate as a Decimal Fraction
Rate to be Applied
Amount of Interest

(1)

( 2)

( 3)

(4)

( 5)

( 6)

( 7)

3

14

11

.4583

.075

.03437

60.15

4

12

8

.3333

.085

.02833

101. 99

4

21

17

.7083

.090

.06375

38.25

4

23

19

.7917

.055

.04354

348.32

14

33

19

.7917

.075

.05938

103.92

The first output of GENIE is the calculating procedures expressed in Level Two Language, and that
is shown in Table 4. To confirm this sequence of
specifications for the calculating. rule, the info~­
mation from one numerical example IS also shown SIde
by side.
5. Third Illustrative Problem,
Advertising Effectiveness

For a third illustrative problem, we take another
simple business situation. The effectiveness of
certain advertising is to be calculated.
The hypothetical situation can perhaps be made
clearest by showing in a tabulation for worked examples two cases, for two ads in the August and
September issues of the hypothetical magazine Scientific Canadian.
6. Future Dialog with a Businessman

Now let us consider the situation in which a
businessman, whom we shall call Mr. Bernard Brown,
and a salesman, whom we shall call Mr. Samuel Smith,
are talking to each other, and the salesman is seeking to persuade the businessman to become interested
in making an arrangement with Mr. Smith's firm for a
service to provide ordinary natural language pro-

Table 3 of Article
Editing Changes for Present Version of GEN I E

.t

FORM 4: DUE PAYMENTS RECORD

'.

0)

( 2)

(3)

(4)

(5)

Loan
No.

Borrower
No.

Date
Due

Amount of
Interest

Total
ReQayment

101

777

8/1/72

$ 60.15

$1810.15

102

658

7/1/72

101.99

3701.99

103

721

11/15/72

38.25

638.25

104

746

12/15/72

348.32

8348.32

105

777

5/15/73

103.92

1853.92

Before this information is typed on paper tape
and input into the present version of GENIE, two
editing changes need to be made, in order to accommodate the limitations of the present version
of GENIE. One change is sequential renumbering of
COMPUTERS and AUTOMATION for July, 1973

Ed i ting Change No.2

Editing Change No.1
Form No.

Column l'~o.
on Form

Column No.
for GENIE

Month
in Letters

Month
~

1
2
3
4
5
6

Jan.
Feb.
Mar.
Apr.
May
June

1
2
3
4
5
6

1
2
3
constant 24
4
constant 100
5
6
7

7
8
9
10
11
12
13
14
15

July
Aug.
Sept.
Oct.
Nov.
Dec.

7
8
9
10
11
12

1
2
3
4
5

16
17
18
19
20

2

1
2
3
4
5
6

3

4

21

Table 4 of Article

Table 5 of Article

Calculating Amount of Interest and Total to
be Repa id - Output from GENIE

Calculating Advertising Effectiveness Worked Examples Input to GENIE

0)

( 2)

Actual Output in
Level Two Language

For Comparison:
a Numerical
Exam121e

FORM 1: RECORD OF ADVERTISING

0)
(2)
(3)
(4)
(5)
( 6)
(7)
(8)
(9)
(0)
(1)

(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(20)

OAT
OAT
OAT
OAT
OAT
OAT

105
777
1750.00
7.5
80172
51573

TAB-l (5)
TAB-l ( 6)
(7) 1- (8)
OAT
(9) I (0)
OAT

14
33
19
24
.7917
100
,075
.05938
103.92

(4) I (12)
(1) ':' (3)

(3)

~~

(4)

0)
( 2)
( 6)
(5)
(3) + (5)

105
777
51573
103.92
1853.92

Abbreviations
OAT
TAB

I-

I
+

DATA
Table 1 of
inverse minus, i.e., subtracted from
divided by
mul tiplied by
plus

gramming via GENIE. The time of this dialog is when
GENIE has been adequately developed.
For after all, no idea is very good in the present world unless it can be marketed -- unless the
producer of the goods or the services incorporating
the idea can persuade a logical customer to pay some
money for it. The logical customer may be a business, a foundation, a university, an office of research in the government, a committee, etc.
Under these conditions we can imagine a future
dialog something like the following.

(4)

0)

( 2)

(3)

Media

Month

Scientific
Canadian
Scientific
Canadian

August

H8

Sept.

HlO

(5)

Key of
Space in
Advertisement Pages
Cost
~

$1100
2000

FORM 2: INCOME AND RESPONSE RECORD
(3)

0)

(4)

Number of
Key
Total
Number of
of AdverOrders,
Inquiries
Income
tisement End of 4 Weeks from Orders End of 4 Weeks
H8

26

$3900

140

HIO

50

7500

188

FORM 3: ADVERTISING MERIT
(1)
(2)
(3)
(4)
(5)

Key of Advertisement
Expected Orders from Inquiries
Expected Order Income from Inquiries
Expected total Gross Income from Advertisement
Factor of Gain or Advertising Merit
(4)

(2)

( 3)

H8

35

$5250

$ 9150

8.31

HlO

47

7050

14550

7.28

(1)

( 5)

Salesman: Yes, that used to be true. But we have
found a way of using ordinary natural language plus
calculation layouts plus worked examples. In the
old days, clever human programmers deciphered what
was intended and told the computer in computer language. Now we have a computer program that deciphers better than a human being can, in many cases.
Businessman: Suppose the information I give you
is incomplete? Suppose there are cases I have not
told you about, or cases that I have forgotten
about?

Salesman: Mr. Brown, if you buy our service, here
is what you get. You tell us specifically all the
kinds of ordinary application problems you want to
run on your computer whenever they come up, and we
give you the programs. You tell us in ordinary
natural language with worked examples, so that you
and we both know exactly what you mean. We have a
clever computer program at our head office that
takes in ordinary natural English and makes compu ter programs. We call it GENIE, because it "hears
and obeys".

Salesman: Two things can happen. One is that
the cases not provided for, not covered, are noticed
by GENIE, and so it asks you questions -- in the
same way as an intelligent clerk could ask you a
question such as, "Mr. Brown, what is the largest
value of goods received that you are willing for me
to sign for?" The other thing that can happen is
that GENIE will not know what you have not told it,
and will therefore make mistakes. But you have just
the same situation with a clerk working for you: the
cases where he has not been told and which he cannot
think of inquiring about, he cannot handle correctly.

Businessman: Mr. Smith, how is that possible?
Everybody has told me that in order to run a computer I have to hire programmers or buy software.
In fact, I am told they cost more than the computer
does.

Businessman: Suppose that I want to make changes
in a procedure. Suppose I have a computer program
that will allocate our expenses one way, and then as
of next fiscal year, I want to change my allocations?

22

COMPUTERS and AUTOMATION for July. 1973

Table 6 of Article

Table 7 of Article

Calculating Advertising Effectiveness Manager's Instructions to a Clerk Input to GENI E

Calculating Advertising Effectiveness Editing Changes for Present Version of GENIE

1. You receive a notice of advertising from the
Advertising Department. From this, you copy, in
this case, into Form 1, the Record of Advertising:
Media, Scientific Canadian; Month of Advertising,
August; the Key to the Ad, H8; the space reported in
magazine pages, ~; and the cost, $1100.
2. Four weeks later you receive for this ad from
the sales managerVs office: the "Number of Orders,
end of four weeks," in this case, 26; and the "Number of Inquiries, end of four weeks," in this case,
140.
3. You copy this information into Form 2, the
Income and Response Record: 26 into Column 2; 140
into Column 4.

Editing Change No.

Edi ting Change No.2
Column No.
for GENIE

Column
Form No.
1

2

~

1

1

2
3
4
5
1
2
constant
3

2
3
4
5
6

4

1
constant
2
3

3

4. Now compute Column 3, "Total Income from Orders: as Column 2 times $150, the typical order
amount.

4

5

5. For Form 3, Estimated Advertising Merit: Column 2, "Expected Orders from Inquiries," is equal
to Column 4 of Form 2, namely 140, multiplied by
0.25, "Expected Conversion Rate," resulting in 35.

Term

~

Scientific
Canadian
August
September
H8
HlO

7777

7

8
9
10
11
12
13
14
15
16

Table 8 of Article
Calculating Advertising Effectiveness Output from GENIE

6. Now compute Column 3, "Expected Order Income
from Inquiries," as Column 2 times $150 average
order, which results in $5250.

0)

(2)

7. Add this amount to "Total Income from Orders,"
Column 3 of Form 2, $3900, and put the result $9150,
in Column 4 of Form 3.

Actual Output in
Level Two Language

For Comparison,
a Numerical ExamQle

8. Compute the "Estimated Factor of Gain" from
the ad, Column 5, which equals Column 4, $9150,
divided by Column 5 of Form 1, $1100, producing in
this case 8.31.

0)
( 2)
(3)
(4)
(5)

OAT
OAT
OAT
OAT
OAT

7777
8
3108
.5
1100

( 6)
(7)
(8)
(9)
(0)

(3)
OAT
OAT
(7) * (8)
OAT
(3)
OAT
0) ,~ (12)
(8) ,~ (3)
(9) + (4)

3108
26
150
3900
140
3108
.25
35
5250
9150
8.31

Salesman: You tell us the modifications that you
want to make in ordinary natural language, and you
give us the old program in computer language. We
take the modifications in ordinary natural language
(with worked examples) and the old program, and we
put them into GENIE and out comes the modified program.
Businessman: I don't believe it.
Salesman: Well, you don't have to believe it on
my say so. Come over to our field office next week
and try running your problem on our computer. By
that time I can have your old expense allocation
program and your new modifications run through GENIE
and there will be a program sitting there tailored
for you which is supposed to do your work as well as
a trained clerk. And if it is not quite right,
GENIE can fix it. And if you remember something
next week that ought to have gone in, GENIE can take
care of that, too.
Businessman: Do you make a completely new program
every time somebody gives you a new set of manager's
instructions and examples?
Salesman: No. A good deal of the time we have
programming modules that are sitting on the shelf
ready to be used. But we take what you tell us in
ordinary English, put that into GENIE, and GENIE
either picks up the modules or manufactures new secCOMPUTERS and AUTOMATION for July. 1973

8
9
3108
3110

(1)

(2)
(3)
(4)

(5)
(6)

(5) II (5)

tions of programs, and often does both, assembling
the useful pieces.
Businessman: Friends of mine have told me that it
is easy to program in FORTRAN, and that it takes
only a day or so to learn FORTRAN.
Salesman: Yes, many people can learn FORTRAN in a
day. But the FORTRAN programs they write on their
own usually don't run. Nearly all the time at the
start, two or three of half a dozen little fussy details are missing or wrong. For example, one of the
rules in FORTRAN is that a variable
Businessman: What's that?
Salesman: Excuse me, a variable is like a heading
on a column in a report, like "expenses for materi23

als," or "territory of a salesman" - as I was saying, a variable which has only whole numbers for its
values must have a label beginning with I, J, K, L,
or rII, so in FORTRAN you would have to write
JEXPENSES or KTERRITORY.
Businessman: Why that silly rule?
Salesman: I don't know, but that is the way FORTRAN was designed about 15 years ago - and there is
nothing at all natural about that silly detail. But
GENIE takes ordinary natural English as you write
it, and it knows this puzzlement in FORTRAN, and so
GENIE and not you has to remember that silly detail.
Businessman: What sort of programs to make for me
could I ask you for?
Salesman: Any business application that you want
to do and are now doing, or that you would like to
do and are not now doing - provided that what you
give me in rules, and instructions, and worked examples, and calculation layouts, is definite and
clear to a clerk. In other words, the information
has to be all there. It has to be down on paper
somehow, and in a style which can be understood by
an ordinary clerk. For example, you are not supposed to use French or German, and you are not supposed to leave out some vital detail, and you are
not supposed to write statements which are wrong or
ambiguous - although the worked examples often resolve ambiguities.
Businessman: What about some specific areas of
application?
Salesman: OK, here is a list of over 40 applications. [See Table 7 in Part IJ
Businessman: Where does that list come from?
Salesman: Out of a "Computer Directory and Buyers' Guide" that lists 2300 applications of computers.
Businessman: Can I use your GENIE program to set
myself up in all those ways?
Salesman: I recommend against it. I think a business would have mental and physical indigestion if
it tried. Instead, anything that you are now doing
usefully under any of these headings - and anything
that you figure out how to do as the months go by can be handled by GENIE, could be translated into a
computer program. But remember, garbage in, garbage
out. A good system for a business depends upon a
lot of fitting to the real world - not on computers
at all.

at the beginning is to stick to clerical systems you
have already proved out for yourself, and translate
those into a computer system. And this is why GENIE
is so remarkable and so important - it cuts the
costs of plain ordinary programming so greatly. On
applications programming it is a breeze and a headache remover.
Businessman: Does GENIE dispense with all other
computer programming?
Salesman: No. There are a number of situations
in regard to which we have so far made no effort to
apply GENIE. Examples are: scheduling and optimizing the operation of computers and peripherals;
chess-playing; optical character reading; problems
already solved by efficient computer programs; problems for which the solution is very hard to express
in ordinary natural language instructions that a
clerk can carry out; etc. We do not see theoretical
barriers to GENIE for interpreting natural language
used to express calculating rules; but at the beginning it is not sensible to bite off more than one
can chew. The logical area to begin with seems to
be applications in business where calculating rules
are being applied by clerks.
Businessman: How did you all work it out?
Salesman: It has been in the wind really. There
had to be some important way around most of the bottleneck of programming. And we think we have found
it, by taking a very broad look at the whole problem, and using all the angles together.
0

CORRECTION

In the article "The Path to Championship Chess by
Computer" by Professor Donald Michie published in
the January 1973 issue of "Computers and Automation," the following corrections should be made:
Page 7, second column, third paragraph, lines
and 2: replace "the Rand Corporation mathematician" by "the professional philosopher".
Page 8, the chess position for Figure IB
should be:

1b

Businessman: So any information handling that I
am doing now, using clerks or using executives, can
go into GENIE?
Salesman: Using clerks, yes; using executives,
probably not. Executives use judgement that is not
usually written down in rules. What you cannot get
into some kind of writing plus some kind of figures
plus some kind of forms probably can't go into any
computer - at least at this time.

Page 9, the graph in Figure 2 should be
as follows (containing one more connecting link):

Businessman: You seem to know a lot about what
can and cannot go into a computer.
Salesman: Well, it happens that I have been in
the computer field for over thirty years. And I
have seen lots of people make a lot of trouble for
themselves with a computer. A good solid guideline

24

COMPUTERS and AUTOMATION for July, 1973

Applications of Comput'ers
in Correspondence Schools
Major Charles Ehin
Qtrs. 611-8 7th St.
Maxwell AF8, Ala. 36113

'The use of on-line computer terminals is due much greater attention by the
administrators of correspondence study institutions with versatile third and
fourth generation systems becoming more readily accessible."

The absence of research and literature dealing
specifically with computer applications in correspondence instruction administration and the lack of
general knowledge about computer systems currently
being used by some home study organizations prompted
the author to conduct an extensive analysis of this
area. l Only the details of the questionnaire survey
phase of the study are presented in this article. 2
Method

All of the sixty-three members of the Independent
Study Division of the National University Extension
Association (NUEA) and the seventy-five members of
the National Home Study Council (NHSC) were selected
as targets for the questionnaire survey. The two
groups were chosen on the assumption that members of
the Associations would respond more readily to the
survey and furnish the most reliable and accurate
information.
Table 1
Status of Computer Applications
by Institutions Surveyed

Institutions

% of NUEA

% of NHSC

Members

Members

Combined
Percentage

Operational

22

44

33

Planned

12

8

10

Status

Of the 138 schools selected for the study, 121 or
88 percent responded. Questionnaires were returned
COMPUTERS and AUTOMATION for July, 1973

by 94 percent of the NUEA member departments and 83
percent of the NHSC member schools. The high rate
of participation was partially attained by sending
out 46 follow-up notices.
Results

As indicated by Table 1, 13 (22 percent) of the
NUEA departments and 27 (44 percent) of the NHSC
schools are using computers to varying degrees in
the administration of their respective programs. In
addition, 12 percent of the NUEA members and 8 percent of the NHSC members have definite plans for incorporating EDP systems into their operations.
Active enrollments of the university affiliated
departments utilizing computers range from 2000 to
12,500 and those not using computers range from 110
to 22,000. The institution with 22,000 enrollments,
however, is in the final proposal stages for installing an EDP system with one of the leading computer manufacturers. The largest NUEA member which
does not have immediate plans for computer applications carries 17,500 active enrollments.
Of the 62 private home study schools responding
to the survey ten declined to indicate the number
of active enrollments monitored by their organizations. With that exception, private schools using
EDP systems have enrollments ranging from 85 to
400,000. The range of enrollments of schools not
using computers is from 80 to 100,000. Interestingly, the organization with 100,000 enrollments has
no immediate plans for incorporating a computer system to assist in the administration of its programs.
The second largest school with 40,000 enrollments
similarly has no immediate intentions of adapting an
25

EDP system to its operations. It should also be
noted that 26 percent of the NHSC members that utilize computers employ the systems only for the maintenance of student financial records.
Specific Modes of Operation

The data in this section has been compiled from
the forty questionnaires returned by the institutions that are using computers to varying degrees
in the administration of the programs.
Methods of Systems Access

In order to gain access to a computer to realize
its benefits an organization must own, lease or
share a system or use any combination of the three
methods.
Table 2 shows the manner in which correspondence
study schools obtain computer time for their data
processing needs. Al though several indeQendent study
departments lease peripheral devices, they all share
CPU time with other university departments and divisions. The procedure is logical since most campuses
have a central computer facility which attempts to
support all data processing requirements.
Table 2
Methods of Gaining Access to Computer System/s and
Frequency of Processing by Institutions Using Computers

Institutions
Method and
Frequency

%of NUEA

%of NHSC

Members
03F

Members
(27)

Combined
Percentage

Method

Own
Lease
Share
Other
Frequency
Daily
Biweekly
Weekly
Bimonthly
Monthly
Other

7

5

33

23

56

70

4

2

15

59

45

15

11

13

24

30

28

Modes of Sharing EDP Systems

There are three primary ways in which to share a
computer system. First, an organization can have
its own keypunch equipment by leasing or purchasing
it. In this manner source document conversion can
be performed on the premises and data manipulation
at a shared computer facility. Second, an institution may have both its keypunching (or another
method of data transformation) and processing requirements accomplished at a service bureau or computer center. Such a method is more appropriate for
smaller agencies with moderate information handling
needs. Third, an institution could purchase or
lease one or more on-line terminals tied into a
shared computer system located at any distance from
the physical facilities of the organization. Input/
output of the terminals is either displayed in
teletype form or on cathode-ray tubes (CRT's). Depending on the capabilities of the central EDP system, the third mode of sharing a computer is the
most responsive and anthropocentric, giving all
users the pretense that they have individual control
of the entire system.
Table 3
Methods of Sharing Computer Systems by
I nstitutions Using Computers

%of NUEA %of NHSC Combined
Members

8

2

15

5

23

7

* The figures in parentheses represent the actual
number of the population under consideration.
**In some cases departments lease card devices or
terminals but they all share a central computer
system.
One extension division leases its own
computer which it shares with the independent
study department.

More than half of the private home study schools
also share EDP systems. This is accomplished by the
use of service bureaus, sharing a central system by
branches located throughout the country, sharing a
system with other affiliated resident schools, or in
one instance utilizing the computer of a local university. In addition, one-third of the NHSC member
organizations gain access by leasing computer systems.
The frequency with which data is processed is
also depicted by Table 2. Obviously, the private
institutions update their computer files more fre26

quently than the university departments. This disparity seems to be partially caused by the fact that
the private home study schools are more involved
with financial transactions than are the independent
study departments. On the other hand, NUEA member
institutions utilize EDP systems more for the generation of statistical summaries than for the support
of day-to-day operations. They therefore require
less frequent processing for the maintenance of
their computer files. According to the survey, 69
percent of the university departments and 59 percent
of the private schools employ computers to produce
course-instruction-related statistical information.

(3)

Members
(27)

Percentage

In-house keypunch equipment with processing. done
at a shared computer center
23
20
21
Keypunching and processing done at a shared computer center
73
68
62
In-house terminal/s tied into a shared computer
center for remote I/O
7

15

Other

7

4

Table 3 gives a breakdown of the methods used for
sharing computer systems and the frequency with
'which they are employed. The NUEA and the MISC members utilize the second mode of shared systems configuration most often. It appears that both groups
should give more consideration to the introduction
of on-line terminals to obtain more adequate management information. This would be especially
appropriate for very small institutions that have
not even considered computer applications.
Storage Media Used for Major Files

The importance of storage medium selection in
the design of data files cannot be over-emphasized.
The critical aspects are the speed of data transfer,
COMPUTERS and AUTOMATION for July. 1973

the arrangement of individual records, and the capacity of the medium. Punched card files are cumbersome, organized sequentially, and have an extremely slow computer input rate. Records on magnetic tape are arranged sequentially but data transfer is fast. Magnetic disk files can be randomly
accessed and they have a very fast data transfer
rate,
The types of storage media used for major files
by the surveyed groups is depicted by Table 4. NUEA
members use primarily punched cards or a combination
of punched cards and magnetic tape for the maintenance of their active enrollment file. The high
utilization of punched cards for this file by the
universities could be attributed to the extensive
use of card devices for statistical compilations by
the departments prior to the introduction of EDP
systems. The use of business machines for such purpos3s was well publicized in 1953 by Mary L. McPartlin of Loyola University. As a matter of fact, in
some instances lesson and examination grades are
posted manually onto punched cards. When a combination of punched cards and tape is used the card file
is usally updated on a daily basis while the tape
file is processed less frequently.
NHSC members prefer to maintain their active
student file on tape or a combination of manual
cards and tape. As stated previously, 26 percent of
the private schools maintain only the enrollees' financial records on computer accessible files. Educational records in these organizations are primarily
kept on manual cards or in folders. A large number
of private organizations, however, utilize magnetic
tape for all active student records.
Table 4
Types of Storage Media Used for Major Files
by Institutions Using Computers
Institutions

%of NUEA %of NHSC Combined
Members

Members

(3)

(27)

File and Medi um

,
I

Active Student File
Manual cards
Punched cards
Magnetic tape
Disk
Manual cards and tape
Punched cards and tape
Manual cards and di sk
Other
Inactive Student File
Manual cards
Punched cards
Magnetic tape
Disk
Microfilm
Manual cards and
tape or disk
Manual or punched
cards and microfi 1m
Other
Inventorl Control File
Manual cards
Punched cards
Magnetic tape
Disk
Combination of any
four above
No records maintained

Again refering to Table 4, it can be observed
that numerous NUEA affiliates prefer a combination
of manual or punched cards and microfilm for the
maintenance of transcript or inactive student files.
The utilization of alternative means is quite evenly
distributed among the remaining departments. NHSC
members on the other hand, use more manual cards for
the storage of inactive student records than any
other type of media.
Manual cards are utilized primarily for the maintenance of inventory control records by both groups
surveyed. The choice of such a medium by the NUEA
members does not seem inappropriate with enrollments
not exceeding 12,500 by the departments employing
computers. For several of the larger private
schools, however, with up to 50,000 enrollments and
available EDP systems, the procedure appears questionable.
Modes of Active Student File Input and Access
The methods of computer data input and information retrieval are directly related to the storage
media employed. As indicated previously, both
groups surveyed utilize punched cards and magnetic
tape extensively for the maintenance of active enrollment files. The addition of records to a
punched card file can only be accomplished by producing new cards and physically or mechanically
placing them into the file. Conversely, the deletions of records kept in a punched card file can be
performed only by manually or mechanically extracting individual cards from the file. Input or changes
to sequential magnetic tape files can be initiated
by the use of punched cards, paper tape, optical
mark reading (OMR) equipment, or optical character
recognition (OCR) units. The latter two devices
have the capability of converting source data directly to tape or disk.
Table 5
Modes of Active Student File Input and Access
by Institutions Using Computers

Percentage

Institutions

%of NUEA %ofNHSC
7
46
8
8

11

7
26
7
26

23
8

8
15

15
15
8

10
20
20
8
17
8
7
10

8

44
4
15
4
7

35
8
12
3
8

8

19

15

38
8

7

12
7

54
15

63
4
11
4

60
7
7
3

8
23

15
3

13
10

COMPUTERS and AUTOMATION for

July.

1973

Members
(3)

Members
(27)

Combined
Percentage

Input
Punched cards
On-line terminal/s
Other

85
15

89
7
4

88
10
2

Access
Punched cards
On-line terminal/s
Manual
Periodic printouts
Other

69
8
23

48

55
10
25
8
2

Input and
Access ~lodes

11

26
11

4

Table 5 indicates that all correspondence study
institutions employ punched cards almost exclusively
to update the active student file. This is logical
considering the primary storage media used by the
agencies and the almost total absenc'e of OMR and OCR
equipment. According to Table 5 much of the information from the enrollment file is also retrieved by
means of punched cards. One fourth of the data,
however, is manually accessed. Most of this may be
attributed to two factors. One reason is that
punched card files of the NUEA departments are usually manually searched for information. Another influence is the fact that numerous NHSC members only
maintain student financial data on computer accessible files.
27

Student Inquiries and Reminder Notices

The institutions surveyed that use computers were
asked whether replies to all student inquiries are
manually prepared. Table 6 depicts the manner in
which answers to student inquiries are produced. Approximately 38 percent of the NUEA members and only
15 percent of the NHSC affiliates respond manually
to correspondence from students, Form letters are
most frequently used by the universities, while most
of the replies by the private organizations are prepared by the combined use of automatic typewriters
and computers.

must also be noted at this juncture that 63 percent
of the private schools, as compared to 8 percent of
the independent study departments, use computer
printed address labels for sending initial course
materials to students. Production of labels is a
simple by-product of computer enrollment processing
and is very useful both for mailing study materials
and inventory control
Table 8

Extent of Computer Application in Test Scoring,
Feedback, and Analysis by
Institutions Using Computers
Institutions

Table 6

Methods Used in Responding to Written Student
Inquiries by Institutions Using Computers
Institutions

%of NUEA %of NHSC Combined
Members
(3)

Members
(27)

Percentage

All manually typed
or written

38

15

23

Some form letters

54

22

33

33

22

15

10

15

12

Methods

%of NUEA
Members
(3)
8':;

cati ons
Test scoring
Test feedback
AQ~li

Test Analysis

Some other combination
of the above methods

8

It should be noted that the use of form letters
is perhaps the most impersonal way of responding to
queries. On the other hand, computer printed replies and especially correspondence produced by
automatic magnetic tape or card controlled typewriters can be made as unique and personal ~s manually composed letters. In most cases such correspondence is more informative and fluent than routinely prepared replies by clerks or counselors.
The NUEA members, therefore, should investigate the
possibilities of taking advantage of the above methods in the future.
Table 7

15
13

Computer Scoring and Posting of Grades

As is indicated by Table 8, almost no university
associated departments and very few private organizations employ EDP systems in scoring lesson tests
or examinations. The obvious reason for this is the
fact that constructed response questions are used
exclusively for lesson tests by the university departments and in most instances by the NHSC members.
All of the private schools that do use computers for
lesson or examination scoring also utilize the systems for the generation of test result feedback for
the students. The feedback includes specific correcti ve conunents or indicates areas to be re-studied.
No conclusive evidence has been presented indicating that essay-type tests are superior to multiple choice response examinations. A study program
cannot be assumed to be personalized simply because
tests are evaluated by persons instead of machines.
Machines are constructed by man and can be as effective as the professionals who design and program
the devices.
Table 9

Extent and Methods Used for Posting Dates of
Lesson/Examination Receipts to Active
Student File by Institutions Using Computers

Institutions
Members
(3)

Members
(27)

Percentage

23

44

38

23

19

20

Extent and Methods

8

74

53

Extent

Table 7 shows the percentage of correspondence
study institutions utilizing various forms of computer generated student reminders or notices. Apparently the NHSC members are applying information
technology more fully in this area, although most of
their effort is centered around delinquent accounts.
There is much that can be accomplished by all correspondence instruction institutions concerning computer-based motivational subsystems development. It
28

22

One NUEA member school uses a computer for only
final examination production and scoring.
-- One NHSC member institution uses a computer for
scoring some tests and examinations.

%ofNUEA %ofNHSC Combined

Delinquent account

Percent age
18

19

Types of Computer Generated Student Reminder
Notices Utilized by Institutions Using Computers

Cancellation

Members
(27)
22':":;

-

Some form letters and
automatic typewriter
replies
Some computer
printed responses

TYQes of Notices
Extended inactivity

%ofNHSC Combined

Institutions

%ofNUEA %of NHSC Combined
Members
(3)

Members
(27)

Percentage

46

19

28

67

100

82

Methods
Keypunchi ng a card for
each test received
Student submi ts a prepunched card with
each lesson
On-line terminal
input after lesson
receipt

17

9

16

9

COMPUTERS and AUTOMATION for July. 1973

u
E

Test analysis is facilitated by the use of computers. As lessons or examinations are evaluated
by an EDP system a test analysis file can be updated
at the same time and statistical summaries can be
prepared periodically or by request. Such a vital
educational tool could and should be continuously
used to assure that lesson tests and examinations
are well structured and accomplish desired objectives effectively. No university departments, and
only 19 percent of the private schools, utilize EDP
systems for test analysis purposes.
When lessons or examinations are received from
the students for instructor grading 46 percent of
the NUEA members and 19 percent of the NHSC associates that use computers post the dates the tests are
received to the active enrollment file prior to distributing the tests to the instructors. According
to Table 9, most of the institutions employ punched
cards for posting the dates of test receipts to the
studen't records, After the lessons or examinations
have been evaluated by the instructors, punched
cards are also the primary means used by all institutions to post grades to the active enrollment
file. As indicated by Table 10. however, 37 percent
of the grades are entered manually to the student
records after instructor notification. Again, this
is largely due to the high rate of manual manipulation of punched cards by the university departments
and the frequent separation of educational and financial student records by the private schools.
Table 10
Modes Employed for Posting Instructor Graded
Test/Examination Results to the Active Student File
by Institutions Using Computers

plans for incorporating a system in the near future.
It appears that many schools may be still using procedures and methods in the administration of their
programs simil~r to the ones described by Arthur J.
Klein in 1919.
The survey has revealed that correspondence instruction departments or schools do not have to be
very large to realize benefits from EDP systems.
Schools with enrollments as low as 85 are using
computer systems in the management of tneir programs. These benefits can, of course, only be attained by sharing computer systems. Specifically,
the use of on-line computer terminals is due much
greater attention by the administrators of correspondence study institutions with versatile third
and fourth generation systems becoming more readily
accessible.
Footnotes

1. Charles Ehin, ·"Analyses and Al ternati ve Proposed

General Designs of Computer-Based Correspondence Instruction Systems," (Unpubli shed di ssertation, University of Oklahoma, May, 1972).
2. See Charles Ehin, "Computer Applications in Correspondence Education," Education Technology,
forthcoming, for a brief summary of the study
and a description of one of the proposed general systems designs.
3. Mary L. McPartlin, "The Use of Business Machines
in the Administration of Correspondence
Courses," Proceedings of the International Conference on Correspondence Education, 1953.
4. Arthur J. Klein, "The Administration of Correspondence Study Dep2rtments of Universities and
Colleges," Bureau of Education Bulletin No. 56,
1919, pp. 3-540
[J

Institutions

%of NUEA %of NHSC Combined
Members

Members
(27)

Percentage
Keypunching a card as a result of instructor
noti fi cati on
62
50
54
Optical mark reading equipment reads instructor
input
5
3
On-line terminal input after instructor
notification
(3)

7

COMPUTER DIRECTORY
AND BUYERS' GUIDE, 1973
- the midyear issue of Computers and Automation,
over 160 pp.
- will be off the press about August 10
The COMPUTER DIRECTORY is:
•

3

Generation of paper tape after instructor
noti fi cati on
4

3

Manual posting after instructor notification
31

THE 19th ANNUAL EDITION OF THE

41

an annual comprehensive directory of the firms which
offer products and services to the electronic computing
and data processing industry

• a basic buyers' guide to the products and services available for designing, building, and using electronic computing and data processing systems - and much more
PRICE

37

Discussion

•

Prepublication price for subscribers to Computers and
Automation whose present subscription does not include
the Directory (magazine address label is marked *N)
. . . . $9.00

• Special prepublication price for non-subscribers
. . . . . $14.00
(price after publication
. . . . . $17.50)

The compiled data indicates that the level of EDP
~ystems utilization in this area, with very few exceptions, is at the mid-second generation computer
development stage. For instance, only one NUEA and
two MISC member institutions are using disk storage
for active student record files. A point to remember here is that third generation systems were first
introduced in 1964.

The Directory is included in the $18.50 full annual subscription (13 issues) to Computers and Automation
(magazine address label is marked *D)

The figures are not much more encouraging when
the extent of EDP systems application is considered.
Only 33 percent of all the schools surveyed are using computers for various purposes (some only for
financial records) and 10 percent have definite

RETURNABLE IN 7 DAYS IF NOT SATISFACTORY
FOR FULL REFUND
Why not take a look at it?

COMPUTERS and AUTOMATION for July, 1973

Send prepaid orders to:
before August 10

CDrnl?.Y~~~
815 Washington Street
Newtonville, Mass. 02160

How can you lose?

29

The Right That Guards
All Other Rights
John Gardner, Chairman
Common Cause
2030 M Street, N. W.
Washington, D. C. 20036

For two and a half years now, Common Cause, with
its more than 200,000 members, has been fighting
pitched battles in the public interest -- and often
winning. We've learned a lot of lessons, but none
more fundamental than this: As far as citizens are
concerned, all their other rights depend on their
right to know. It's no accident that the First
Amendment comes first: if freedom of information is
successfully stifled, all the other freedoms can be
snuffed out at leisure. What citizens don't know,
they can't object to.
The good investigative reporter is, therefore, an
invaluable ally when citizens are fighting battles
in the public interest. And reporters are virtually
unanimous in asserting that agressive investigative
reporting could not survive without extensive use of
confidential sources. Such sources will fall silent
if it becomes apparent that courts can force reporters to reveal their identity.
Unfortunately, courts and prosecutors allover
the country seem to be seeking just that result. In
the last several years, journalists have repeatedly
been subpoenaed to reveal confidential information,
and some who refused to talk have gone to jail.
Present safeguards are inadequate to protect the
public's right to know. The Attorney General's
guidelines for the issuance of subpoenas to the news
media apply only to the Department of Justice, contain significant loopholes, and may be changed at
the whim of the Attorney General. Eighteen states
possess newsmen's privilege statutes, but these laws
are usually narrow and imprecisely drafted, and
in many cases have been gutted by judicial
interpretation.
In June, 1972, a narrowly divided Supreme Court
ruled 5-4 in Branzburg v. Hayes that a newsman had
no First Amendment right to refuse to reveal his
confidential sources to a grand jury. The court emphasized, however, that Congress or state legislatures were free to enact newsmen's shield legislation, and it virtually invited such action.
The public's right to know should not depend on.
the willingness of newsmen to go to jail rather than
betray confidential sources. A shield law will benefit the public as well as newsmen and is not really
a "privilege". Protection for sources is a necessity as basic as a typewriter if a journalist is to
search out the truth. Congress should act now to
protect the free flow of information to the public.
A federal shield law should apply to reporters
for newspapers, magazines, television and radio, and
should protect the identity of informants and the
information they give in confidence. It should apply ~ithout exception to grand juries, legislative
hearings, and administrative proceedings. It should
cover all civil trials with the possible exception
of libel cases. This broad coverage is necessary to
guard against "fishing expeditions" made possible by
the lack of evidentiary rules at investigatory proceedings and the broad discovery possible in civil
trials.
30

Criminal trials present a different problem, and
here there should be a narrow and well-defined exception. The scope of inquiry here is sharply focused since a jury or prosecutor has formally preferred charges and found probable cause to believe
that the defendant has committed a crime.
Furthermore, to make an effective defense, a
criminal defendant may have a compelling need for
confidential information possessed by a reporter.
Any newsmen's privilege that Congress enacts would
have to give way, in any case, before the Sixth
Amendment's guarantee to a criminal defendant of
compulsory process for obtaining witnesses and for
confronting the witnesses against him.
The exception allowed in criminal trials must be
extremely specific. It should not allow a court to
engage in a general "balancing of equities" that
would make it difficult for informants and reporters
to predict the situations in which their agreements
of confidentiality might be invaded. Three conditions should be met before disclosure is required:
1. the trial should involve a specified serious
crime;
2. there should be a demonstration of probable
cause to believe a newsman possesses relevant information; and
3. it should be shown that no alternative means
of obtaining the information exists.
Enactment of a statutory newsmen's privilege will
never prevent a reporter from testifying if he feels
his obligations as a citizen compel him to do so.
In cases involving immediate danger to human life,
such as planned bombings, kidnappings, hijackings,
and incidents of food poisoning, reporters have in
the past willingly presented whatever knowledge
they have to public authorities regardless of
confidentiali ty.
No doubt a statutory privilege will sometimes operate to exclude trustworthy and relevant evidence
from judicial proceedings, but this is justified by
the central role played by the press in the preservation of free self-government. Thomas Jefferson
said, "The basis of our government being the opinion
of the people, the very first object should be to
keep that right; and were it left to me to decide
whether we should have a government without newspapers or newspapers without a government, I should
not hesitate a moment to prefer the latter."
ADVERTISING INDEX
Following is the index of advertisements. Each item contains:
product / name and address of the advertiser / name of the agency, if
any / page number where the advertisement appears.
COMPUTER-ASSISTED ANALYSIS AND DOCUMENTATION OF
COMPUTER PROGRAMS / Berkeley Enterprises, 815 Washington St.,
Newtonville, Mass. 02160 / page 50
COMPUTER DIRECTORY AND BUYERS' GUIDE, 1973/ published by
Berkeley Enterprises, Inc., 815 Washington St., Newtonville, Mass.
02160 / page 29
COMPUTERS AND AUTOMATION / Computers and Automation, 815
Washington St., Newtonville, Mass. 02160 / page 52
THE FLY, THE SPIDER, AND THE HORNET / taken from Ride the
East Wind: Parables of Yesterday and Today, published by Quadrangle
Books (a New York Times Company) / Computers and Automation,
815 Washington St., Newtonville, Mass. 02160 / page 39
THE NOTEBOOK ON COMMON SENSE, ELEMENTARY AND ADVANCED / published by Computers and Automation, 815 Washington
St., Newtonville, Mass. 02160 / pages 2, 3
WHO'S WHO IN COMPUTERS AND DATA PROCESSING / jointly published by Quadrangle Books (a New York Times Company) and
Berkeley Enterprises, Inc., 815 Washington St., Newtonville, Mass.
02160/ page 51

COMPUTERS and AUTOMATION for July, 1973

The American News Media and the
Assassination of President John F. Kennedy:
Accessories After the Fact 1 - Conclusion
Richard E. Sprague
Hartsdale, N. Y. 10530

'~

. . the pattern of internal knowledge of conspiracy followed by complete suppression of

such knowledge is too strong to ignore."

Part One of thi sarti cle, publi shed in the June
issue of Computers and Automation, dealt with the
interlocking nature of the news media and various
attempts at blunting independent investigations of
the assassination.
The New York Times

The record of the Times through the 1964-1971
period follows the same pattern as the CBS and
Life magazine editorial policies.
The early editorials following the Warren Report
supported the Commission. The Times cooperated by
publishing much of the report in advance. In 1965,
however, editorials began to appear questioning the
Commission's findings and suggesting a new investigation. In 1964 the Times formed a research team to
investigate the assassination, headed by Harrison
Salisbury. The team of six included Peter Khiss and
Gene Roberts. Their conclusions were never made
public by the Times, but indications point to their
finding evidence of conspiracy.
Khiss in particular, through the 1966-68 period
in several meetings and discussions with the author,
expressed doubts about the Warren Report and questioned the lone madman assassin theme. Then, when
the Garrison investigation made the news, the Times
began a regular campaign to undermine Garrison's
case, to support the Warren Commission, and finally
during the Clay Shaw trial to completely distort the
news and the testimony presented. Martin Waldron
was the reporter sending in the stories from the
Shaw trial, but someone in New York edited them to
completely change their content. The author saw the
first day of the trial story as written by Waldron
and the final version appearing in the Times. The
two were completely different, with Waldron's original following the actual trial proceedings very
closely.
The author, writing under a pen name, Samuel B.
Thurston, postulated the possibility that The New
York Times on selected subjects, including the JFK
assassination, is controlled by the CIA through
their representative among top management, Mr.
Harding Bancroft. 2
In the summer of 1968, the author discovered a
remarkable similarity between the sketch of the as1. Accessories After the Fact is the title of a book
by Silvia Meagher published by Bobbs Merrill in
1967, accusing the Warren Commission and the various
government agencies of covering up the crime of the
century. This article accuses the national news
media of the same crimes.
2. Computers and Automation, July 1971.
COMPUTERS and AUTOMATION for July. 1973

sassin of Dr. King and one of three tramps arrested
in Dealey Plaza following the assassination of President Kennedy. Peter Khiss wrote a story about this
and it was published by the Times in June 1968. Apparently that was the final straw for Times management as far as Khiss was concerned. He has not been
allowed to do any more research on assassinations or
to even discuss the subject at the Times since then.
As he told the author in 1969, he doesn't attend any
press conferences about assassinations because he
doesn't like it when people in Times management say,
"Here comes crazy old Pete Khiss again with his conspiracy talk".
The acme of The New York Times actions and editorial positions on the JFK assassination came in
NoverMer and December 1971. They published three
items supporting the Warren Commission eight years
after the assassination, at a time when it seemed to
be a dead issue, at least on the surface.
The first was a story about Dallas eight years
later by an author from Texas who wrote his entire
story as though it were an ~stablished fact that Oswald was the lone madman assassin firing three shots
from the sixth floor window of the Depository building and later killing policeman Tippitt.
The second was an Op Ed page guest editorial by
none other than David Belin, Warren Commission lawyer. He defended the Commission and attacked the
researchers. The third was a story by Fred Graham
about the findings of Dr. Lattimer who was allowed
to see the autopsy photographs and X-rays. Graham
actually wrote most of his story which solidly
backed up the Warren Commission due to Lattimer's
claims that the autopsy materials proved no conspiracy, before Lattimer ever entered the Archives.
In other words, it appears that Graham knew what
Lattimer was going to find and say in advance. Either that, or someone in Washington, D.C. gave someone high up at the Times orders in advance to prepare
the story for page 1, upper left-hand corner, of the
paper. It really didn't make any difference whether
Dr. Lattimer ever saw the X-rays and photos.
The concerted campaign on the part of Times management could have been timed to prevent a discovery
of new evidence of conspiracy in the autopsy materials. The reason for this possibility developing in
the November 1971 period is that the five year restriction placed on the autopsy evidence by Burke
Marshall, Kennedy family lawyer, expired in NoverMer
1971. Four well-known and highly reputable forensic
pathologists, Dr. Cyril Wecht of Pittsburgh, Dr.
John Nichols of the University of Kansas, Dr. Milton
Halpern of New York City and Dr. John Chnpman of Ue31

troit, had already asked permission to examine the
X-rays and photos upon the expiration of the five
year period. All four are known to question the
Warren Commission's find!ngs. What better way to
freeze them out of the Archives than to select a
doctor who could be trusted to back up the Commission (Lattimer had published several articles doing
just that), commission him to go into the Archives,
and then get The New York Times to publish a Sunday
page 1 story demonstrating that no one else need
look at the materials because they back up the Warren Commission's findings completely.
All attempts by researchers including the author
to convince Times management that the other side of
the story should be told have been completely ignored. Lattimer's findings, if correct, actually
prove conspiracy. The Times has been informed of
this, including Fred Graham, Harrison Salisbury (Op
Ed page editor), but they have shut off all discussion on the subject. The complete story of the complicity of The New York Times in the crimes to which
they have become an accessory would take up an entire volume. A researcher 3 is assembling the entire
story and has published it in a recent magazine article.

NBC
The Nati onal Broadcasti ng Company became an acti ve
participant in the government's efforts to protect
Clay Shaw and to ruin Jim Garrison.
Two of NBC's high level management people, Richard
Townley of NBC's affiliate in New Orleans, WDSU, and
Walter Sheridan, executive producer, became personally and directly involved in the Shaw trial. They
were indicted by a grand jury in New Orleans for
bribing witnesses, suppressing evidence and interfering with trial proceedings. The indictments are
still open. NBC top-level management backed Sheridan and Townley totally.
NBC produced a highly biased, provably dishonest
program personally attacking Garrison and defending
Shaw, prior to the trial. Frank McGee who acted as
moderator later had to publicly apologize for lies
told on the program by two "witnesses" who NBC paid
to give statements against Garrison. The FCC ruled
that NBC had to give Garrison equal time because the
program was not a news program but a vendetta by NBC
against Garrison. NBC did give Garrison 30 minutes
(compared to their one hour attack) to respond at a
later date. Sheridan was the producer of the one
hour show.
With Sheridan and Townley so deeply involved, and
with such an extremely strong editorial position
favoring the Justice Dept., the Warren Commission,
and the lone assassin stance, it naturally raises
suspicions about NBC's and RCA's independence and
news reporting capabilities.4 At one pOint in 1967,
the President of NBC, according to Walter Sheridan,
helped in the bribery efforts by calling Mr. Gherlock, head of Equitable Life Assurance Co.'s New
York office, and asking for assurance that Perry
Russo, who worked foi Equitable, would cooperate
with NBC.
NBC is also an owner of several important pieces
of photographic evidence. A TV film taken by NBC
photographer David Weigman was suppressed by NBC and
not made available to researchers. It shows the
grassy knoll in the background just a fraction of a
minute after the shots. Some of the assassination
participants can be seen on the knoll.
3. Jerry Policoff. The Realist, January 1973.
4. A detailed review of NBC's performance and Walter
Sheridan's and Richard Townley's involvement is
given in The Kennedy Conspiracy by Paris Flammonde.
32

Fortunately for researchers, NBC sold the Weigman
film to the other networks and to the news film
agencies before realizing its importance. The author was able to purchase a copy from Hearst Metrotone News.
NBC's affiliate in Ft. Worth, WBAP, has in their
files several important film sequences. James Darnell took several sequences on the grassy knoll and
in the parking lot which should contain important
evidence. Dan Owens took TV movies in and around
the Depository building which should show how the
snipers' nest was faked on the sixth floor, and one
of the assassins in front of the building.

ABC
Of the three major TV networks, ABC has remained
more objective and appears to be less under the
thumb of the government than the other two. For example,. when NBC was busy defending the Warren Commission and Clay Shaw and attacking Jim Garrison,
ABC gave Garrison a free chance to express his views
without interruption on their Sunday program, "Issues and Answers". They have never taken an editorial position one way or the other on conspiracy.
However, in the Robert Kennedy assassination case,
an investigation into evidence of conspiracy was
suppressed at ABC. The man heading the brief investigation was stopped and sent off to Vietnam.
The man at ABC who called the shots in stopping the
inve,stigation, and in suppressing evidence in ABC's
possession, was a lawyer named Lewis Powell.

.1

The evidence owned by ABC is a video tape of the
crowd in the Ambassador Hotel ballroom before, during and after the shots were fired in the kitchen.
The ballroom microphones, including ABC's, picked
up the sound of only three shots above the crowd
noise. Since Sirhan fired eight shots, or certainly
more than three, and since Los Angeles police tests
proved that Sirhan's gun could not be heard in the
position of the microphones in the ballroom, the ABC
film and soundtrack is important evidence of three
other shots, not Sirhan's.
The sequence was originally included in the TV
film of Robert Kennedy's 1968 campaign and assassination, "The Last Journey". Following a meeting
at ABC when the management learned what the film
showed, the next TV broadcast of "The Last Journey"
sCheduled for the following week on ABC was cancelled without any logical explanation. The next
time the film appeared on ABC (late 1971), the three
~hot ~allroom sequence had been cut out.
United Press International

UPI has perhaps come closest to really pursuing
the truth about the JFK assassination of all the
fifteen major news organizations included herein.
Yet they, too, have suppressed evidence, have not
had the courage of their convictions in analyzing
conspiratorial evidence, and by default have become
accessories after the fact.
Two different departments at UPI became involved
in the photographic evidence of the JFK assassination. The regular photo news service department,
which receives wire photos and negatives from many
sources allover the world, accumulated a large coliection of basic evidence both from UPI photographers
and by purchasing wire service photos from newspapers, Black Star, AP and other sources. This department has made all of its photographs available
to anyone at reasonable prices ($1.50 to $3.00 per
print).
UPI photographer Frank Cancellare was in the
motorcade and snapped several important photos. In
addition, five other photos at UPI taken by 3 unCOMPUTERS and AUTOMATION for July. 1973

\

"

known photographers are significant. All of these
were purchased by the author from UPI.
The other department has, however, not been as
cooperative. Within the news department at UPI,
Burt Reinhardt and Rees Schonfeld have varied in
their attitude and performance. UPI news purchased
the commercial rights to two very important films
shortly after the assassination. These were color
private-citizen movies taken by Orville Nix and
Marie Muchmore. Both show the fatal shot striking ,
the President, and both show evidence of conspiracy.
In the Nix film, certain frames when enlarged show
the man who fired the fatal shot, on the grassy
knoll with a rifle. Both movies show a puff of
smoke generated by another one of the men involved
in the asiassination.

·'
"

UPI, under the direction of Burt Reinhardt, did
several things with the Nix and Muchmore films.
First, they produced a book, Four Days, which included several frames from the movies, reprinted in
color. Second, they made a composite movie in 35 mm
size from the original 8 mm movies. The composite
used the technique of repeating a frame several
times to give the appearance of slow motion or stop
action during key sections of the films. Reinhardt,
Schonfeld and Mr. Fox, a UPI writer, made the composite movie available to researchers at their projection studio in New York in 1964-65.
Fox and Schonfeld wrote an article for Esquire
magazine in 1965 which portrayed the Nix film as
proving the conspiracy theories about assassins on
,the grassy knoll to be false. This was deemed necessary by UPI management because a New York researcher and a photographic expert, after seeing the
Nix film at UPI, claimed it showed an assassin with
a rifle standing on the hood of a car parked behind
the knolL
The research team had used a few frames from the
film in color transparencies and enlarged them in
black and white to show the gunman.
In 1964 UPI gave the Warren Commission copies of
both the Nix and Muchmore films for analysis. When
the Commission ceased to exist, the films were
turned over to the National Archives under a special
agreement between UPI and the Archives. This agreement reminds one of the agreements between the Archives and the Kennedy family on the autopsy materials, and the obscure one between Life magazine, the
Commission, the Secret Service and the Archives on
the Zapruder film.

;

•

The UPI agreement prevents anyone from obtaining
copies of the Nix and Muchmore films or slides of individual fr,ames for i!!!Y purpose, commercial or
otherwise. The agreement is just as illegal as the
other two, yet it has been just as effective in suppressing the basic evidence of conspiracy.
In 1967, UPI, apparently still not sure they
would not be attacked by researchers on what the Nix
film revealed, employed Itek Corp. to analyze the
film. (At least it would appear on the surface that
UPI did the hiring.) Itek Corp., a major defense
contractor, did an excellent -job of obscuring the
truth .. In an apparently highly scientific analysis
using computer-based image enhancement, they "proved"
that not only was there no gunman on the grassy
knoll, but there was no person on the knoll at all
during the shooting.
The final Itek report was made public and highly
publicized by UPI. It looked to the lay public as
COMPUTERS and AUTOMATION for July. 1973

though the UPI earlier claim of no gunman had been
scientifically substantiated. As a by-product, Itek
got some great publicity for their commercially
available photo-computer image enhancement system.
What the public did not know was the following:
UPI gave Itek only 35 mm enlarged black and white
copies of selected frames from the Nix film. The
amount of detail lost in going from 8 mm color to
35 mm black and white is very high.
UPI gave Itek frames from the Nix film carefully
chosen to not show the gunman on the knoll.
UPI and Itek defined "the grassy knoll" in a very
limited and carefully chosen way so as to exclude
the five other people (in addition to the fatal shot
gunman) on the knoll, who appear in the Nix film as
well as in every other photo and movie taken of the
knoll ?t the time the shots were fired. 5 In addition, man u2 who had ducked down behind the stone
wall during the Nix film, could not, of course, be
detected by Itek since all they had was the Nix film.
Three men standing on the steps on the knoll, and
two men behind the picket fence were completely ignored or overlooked.
The author began to contact Schonfeld and Reinhardt in early 1967, viewed the two films both at
UPI and in the Archives, and requested copies of the
original 8 mm color films or color copies of individual frames. The response to the requests were
negative for more than four years. During this
time, however, the author, the New York researcher,
and a photographic specialist enlarged in color, the
correct frames from the Nix film. The enlargements
clearly show the gunman, not ~ of the car, but
in front of the car, with his rifle poised. He is
standing on a pedestal protruding from the eightsided Cupola behind the stone wall on the knoll.
The car is parked behind the Cupola and can be seen
in several other photos and movies.
Unfortunately, UPI's agreement with the researcher
prevents making public the enlargements in color.
UPI has consistently suppressed this evidence to
this date. They offered in 1971 to make the film
available for a very large sum of money, but have
never agreed that it shows anyone on the knoll, and
will not make copies available for research.
The UPI'editorial positiOn in articles, the book
Four Days, letters, and news releases, has supported the Warren Commission through the years. The
major difference between UPI and Life or CBS is that
no drastic reversal of management policy took place.
AP

Associated Press became ,a definite accessory
after the fact by taking an action unprecedented
for a news wire service. They "published" a threepart report by three AP writers in 1967, completely
backing up the Warren Commission. The report was
transmitted by wire to all AP subscribers over a
three day period and it occupied a total of nine
to ten full pages of the average newspaper. It was
not news, but editorial policy and position, supporting the Warren Commission and the official
5. Those interested in more detail are referred to
the map in the May 1971 issue of Computers and Automation on the JFK assassination. The UPI definition
of "the grassy knoll" was the area bounded by the
picket fence, the stone wall, the top of the steps
on the south, and the cupola.
33

government propaganda about the assassination of
our President.
As pointed out in the first part of this article,
most small newspapers rely on UP and AP for their
news stories. The AP three-part report ran in hundreds of papers across the United States without opposition commentary. For many this was the gospel
at the time. What more could the conspirators and
their government protectors have asked?
AP photographers were on the scene in Dallas during the killing. James Altgens, one of AP's men assigned to Dallas, took seven important photos in
Dealey Plaza. Henry Burrows, an AP photographer
from Washington, D.C., was in the motorcade and snapped two pictures. Four other AP photographers took
ten important photos" In addition AP's photo department, Wide World Photos in New York, purchased
many other photographs taken in Dealey Plaza.
Meyer Goldberg, the manager of Wide World Photos,
set a policy early in the 1966-1967 period, which
placed AP in the position of partially suppressing
basic photographic evidence. The policy contained
several parts. First, Goldberg made it extremely
difficult for anyone to obtain access to the photographic evidence, particularly the negatives. Second, he set a high enough price on copies of photographs ($17050 for one 8 x 10 black and white print)
to freeze out all but commercially financed interests. Thirdly, when an original negative was discovered (by going around the organizational roadblocks set up by Goldberg), the print order, when
cleared by Wide World, was always cropped. Full
negative prints showing important details in the
Al tgens photos were nearly impossibll,e to purchase.
Fourth, whenever any suggestion was ~ade to Wide
World that their photographs contain~d basic evidence of conspiracy, Goldberg and AP ~anagement
turned blue with anger and literally refused to
discuss the subject or permit research in the files.
Various researchers, including Josiah Thompson,
Raymond Marcus and the author, met this type of stiff
opposition, but after many visits, discovered ways
around it. The working staff at Wide World incharge
of the photographic files were more cooperative, and
at least one staff member was completely convinced
about conspiracy.
Nevertheless, the broadly announced editorial
policy and stance of Associated Press since 1964 and
through 1972 has been full support of the Warren
Commission and the lone assassin fable.
Newsweek

Newsweek editorial policy and reporting about the
assassination and its aftermath became largely a
function of the views of one man, Hugh Aynesworth.
Aynesworth was the Dallas-Houston correspondent for
Newsweek following the assassinationo He had been a
reporter on the Dallas Morning News. He was on the
spot when it happened, and turned in several stories
in the days and weeks immediately following November 22, 1963. His point of view was always closely
allied with the Dallas police, the district attorney and the FBI. He supported wholeheartedly the
Warren Report upon its appearance. Up to 1967 his
performance was really not much different than a majority of other non-inquiring reporters who took
what was fed to them and popularized it.
However, in May 1967, after Garrison's investigation hit the news, Aynesworth wrote and Newsweek
published a violent attack on Garrison's investiga34

tion. Aynesworth accused Lynn Loisel on Garrison's
staff of bribing Al Beaubolf to testify about a
meeting to plot the assassination. Beaubolf later
denied this accusation in a sworn affidavit and
proved Aynesworth and Newsweek to be fabricators of
information just like NBC.
Saturday Evening Post

As in the case of Newsweek, the position of the
Saturday Evening Post solidified after the Garrison
probe became public and was based largely on the antics of one man, James Phelan. Phelan wrote a blistering article for the Post published on May 6,1967.
He attacked Garrison and Russo with a special claim
that Russo's original ~tatement to Assistant D.A.,
Andrew Sciambra, differed from his testimony later
on. In view of the earlier editorial position of
the Post when Lyron Land and his wife, both editors,
questioned the Warren Commission findings, the
Phelan article came as somewhat of a surprise. In
fact, the Post had taken a strong conspiracy stand
when earlier in 1967 they published a long article
excerpted from Josiah Thompson's book, Six Seconds
in Dallas, and featured it on the cover.
The Garrison investigation, however, turned the
Post around as it did many other news organizations.
Phelan became directly involved in the case, and in
a sense, was more of an accessory than Walter Sheridan or Richard Townleyo He travelled to Louisiana
from Texas, spent many hours with Perry Russo and
other witnesses, and generally obfuscated the Shaw
trial picture.
Phelan joined the efforts of Sheridan and Townley to get Russo to desert Garrison and to help destroy Garrison and his case. According to a sworn
Russo statement, Phelan visited his house four times
within a few weeks. Phelan told Russo he was working hand in hand with Townley and Sheridan and that
they were in constant contact and were going to destroy Garrison and the probe. Phelan told Russo
that he should go over to the other side because he
would be the only one who would get hurt as a result
of the trial because Garrison would leave him all
alone standing in the coldo
Phelan offered to hire a $200,OOO-a-year lawyer
from New York for RU$SO if he would cooperate
against Garrison. He asked Russo how he would feel
about sending an ihnocent man (Clay Shaw) to the
penitentiary. Phelan left New Orleans and Baton
Rouge to go to New York from where he telephoned
Russo several times offering to pay Russo's plane
fare to New York to meet with him and discuss going
over to Clay Shaw's side.
Because Phelan's article attacked Garrison, hewas
subpoenaed by Shaw's lawyers during a hearing in
1967. Sciambra welcomed the opportunity to crossexamine Phelan on the stand. He described the article as being incomplete and distorted and tantamount
to lying. Sciambra said, "I guarantee that he (Phelan) will be exposed for having twisted the facts in
order to build up a scoop for himself and the Saturday Evening Post."
--Sciambra went on to say that Phelan had neglected
the most important fact of all in his article. It
was that Phelan had been told by Russo in Baton
Rouge that Russo and Sciambra had discussed the plot
(to assassinate JFK) dialogue at their initial meeting.
The statements by Sciambra were very mild in view
of Phelan's actions.

COMPUTERS and AUTOMATION for July. 1973

Capital City Broadcasting

This organization owns several radio stations in
the capital cities of various states and in Washington, D.C. Their interests in the JFK assassination
increased in 1967 and 1968 when the Garrison-Shaw
case was making headlines. A producer at Capital
City, Erik Lindquist, decided to do a series of programs designed to get at the truth. The author furnished various evidence for scripts to be used in
the programs. After several months of work, the
project was cancelled, presumably by top management,
and the broadcasts never did take place.
No further activities are known to the author on
the part of Capital City.
North American Newspaper Alliance

This newspaper chain, with affiliated papers in
smaller communities throughout the north and east,
followed the same pattern of Warren Commission support as all the other major newspaper services and
chains. The Alliance also became involved in the
Martin Luther King case because they circulated the
syndicated column of black writer and reporter
Louis Lomax. He had taken an intense interest in
finding out what happened in the King assassination.
Lomax located a man named Stein who had taken a
trip with James Earl Ray from Los Angeles to New Orleans. The two retraced the auto trip of Ray and
Stein beginning in Los Angeles and heading through
Arizona, New Mexico and Texas. They were trying to
find a telephone booth somewhere along the route
from which Ray had called a friend named Raoul in
New Orleans. Raoul, according to Ray, was the man
who actually fired the shot that killed King. Stein
remembered that Ray told him he was going to meet
Raoul in New Orleans and phoned him at someone
else's office. Stein couldn't remember exactly
where the phone booth was, because he and Ray had
been driving day and night without stopping.
Lomax wrote a series of articles portraying
Raoul as the killer and Ray as a patsy, and sent it
to the Alliance, one column each day from the places
along the retraced trip where he and Stein stayed.
FinallyJ Lomax's column said they had found the
phone bdoth at a gas station in Texas and that he
was going to obtain the phone number Ray had called
in New Orleans. Presumably, he was planning to
visit the local telephone company office the next
morning to obtain the number.
That was the last Lomax column ever to appear in
the North American Newspaper Alliance papers. He
just seemed to disappear completely. The readers
were left hanging not knowing whether he got the
phone number or whether he discovered who it belonged to. The Committee to Investigate Assassinations located Lomax several months later and asked
him what had happened.
He said he had been told by the FBI to cease and
desist from his investigation and not to publish or
write any more stories about it. He said he found
the phone number and where it was located in New Orleans. He gave the number to the Committee. He said
he was afraid he would be killed and decided to stop
work on the case.
Whether North American Newspaper Alliance management knew about any of this, or whether they sanctioned the stopping of the series, rem~ins unknown.
What is known, however, is that Louis Lomax died in
a very mysterious manner in 1970. He was driving
alone along a highway in New Mexico, travelling
COMPUTERS and AUTOMATION for July. 1973

(,

east, very near the same route he and Stein had traversed. He drove off the highway after travelling at
a very high speed, and was found dead, according to
the State police report. Lomax's wife says he was a
very careful driver and never drove at high speeds.
Dallas Newspapers

The two newspapers in Dallas, The Times Herald
and The Morning News, became accessories after the
fact by suppressing evidence of conspiracy and evidence showing the involvement of members of the Dallas police in framing Lee Harvey Oswald. The management policy of both papers in supporting the official positions taken by the Dallas police and district attorney, the FBI and the Warren Commission,
was not immediately established. For the first few
days following the assassination, both newspapers
printed anything which came along. The editions on
November 22, 23, 24 and 25 make very interesting
reading for the researcher because the stories were
printed before anyone had any idea about what to
suppress.
For example, there are stories about other people
being arrested, about other rifles being found near
Dealey Plaza, and about Oswald's rifle being a Mauser
and a British 303 model.
However, within a couple of weeks, editorial and
management policy took over, and the lone madman assasin story received all the attention from then on.
Through the years, the two papers have not made any
independent inquiries, have not been interested in
any conspiratorial discussion, and have remained
completely faithful to the official position.
There were some inquiring reporter types around
(like Ronnie Dugger, for example, orLonnie Hudkins),
but they were squelched completely by management or
by the FBI and Dallas police. Photographers at the
two papers left town or else were completely frightened out of talking about the case or their photos.
Some of these photographs showed evidence of conspiracy, including pictures of three conspirators
under arrest in Dealey Plaza. Other photos proved
that members of the Dallas police altered and
planted evidence in the Depository building in order
to frame Oswald.
Between the assassination and 1967, the manage- ~
ment and owners of the Herald and News were not com- \
pletely aware of the significance of some of the
\
evidence in their files. Nor were they attempting
to control their reporters and news staff. For example, Hudkins found out that Oswald had been a paid
informer for the FBI. He even found out what his
pay number had been (5172). He took the information
to Waggoner Carr, Texas Attorney General, in January
1964. Carr brought it to the attention of the Warren Commission and after Hoover denied it, the matter died in secret executive sessions of the Commission.
Several photographs taken by Dallas Morning News
photographer Jack Beers proved that the police created the so-called "sniper's nest" from which Oswald
allegedly fired the shots. The pictures show the
positions of cartons in the sixth floor window before the police moved them. Beers' photos also-rndicate that the police made the large paper bag inside the Depository building which was used to help
frame Oswald.
Beers was permitted to use his photos commercially in a book, jointly published with R. B. Denson, called Destiny in Dallas. If it were not for
that event, researchers would probably never have
35

\.

seen Beers' photos. Once the Morning News editor,
Mr. Krueger, discovered that the photographs demonstrated both conspiracy and the complicity of some
of the Dallas police force, he locked them up and
made them unavailable to anyone. The pictures remain suppressed to this date.
The Times Herald's record is not much better. For
a few years, through 1967, John Masiotta, the man in
charge of the assassination photos taken by William
Allen, made copies available on a very limited basi s.
The basis in the author's case was that a total of
twelve pictures out of seventy-three taken by Allen
could be purchased. The author was allowed to examine 35 mm contact prints (about 3/4" x 1/2") of
the rest, and the selection decision was extremely
diffIcult. Three of Allen's photos showed the
"tramps" under arrest who were part of the conspiraCy.
Once Times Herald management realized the implications of some of Allen's pictures in pointing out
the real assassins, they too (in 1968) locked the
files and have not permitted anyone to date to see
the phbtos again or to purchase copies.
One photograph, taken by Dallas Times Herald
photographer Bob Jackson, was apparently so obviously in oppositibn to the official police case,
that it was suppressed by late 1966. Jackson was
riding in one of the news photographer's cars in
the motorcade, with Dallas Morning News photographer, Tom Dillard. As the Jackson car approached
the Depository building travelling north on Houston
Street between Main Street and Elm Street (see map
in May 1970 C&A article), Jackson snapped a picture.
At the time, the Kennedy car was undoubtedly already
around the fatal turn onto Elm Street and was probably close to the position where the first shot was
fired. Jackson's car was eight cars behind Kennedy's, or about twenty car lengths.
Jackson can be seen taking this picture in the
Robert Hughes film and in some of the TV footage
taken by other photographers. He also testified
that he took the picture. When the author asked
Masiotta about the Jackson photo in early 1967, he
became very flustered and claimed to know nothing
about it. Jackson himself was finally located and,
when asked about it, became very angry and denied
taking a picture. That photograph has never been
seen by anyone outside of the Times Herald staff.
The speculation about what it probably showed is not
difficult, since the Hughes film, the Weaver photo,
the Dillard photo, and the Tom Alyea first TV sequence all show the same thing. Jackson's photo,
without doubt, showed "Oswald's window" in the Depository building ~ when Oswald should have been
in it. This was embarrassing to Jackson as well as
to the newspaper because he had testified he saw
someone in that window with a rifle. His own photo
would have proved him a liar. Certainly if Jackson's photo (or anyone else's for that matter) showed
Oswald in the sixth floor window, the whole world
would have heard about it on November 22, 1963.
The Dallas Morning News, the Dallas Times Herald,
Jack Krueger, Jack Beers, Bob Jackson, John Masiotta
and others all became accessories after the fact. It
makes no difference whether they made their own decisions, or whether they were encouraged to suppress
evidence by local authorities.
Ft. Worth Star Telegram

Shining like a light in the Texas darkness, the
Ft. Worth Star Telegram did not suppress the evidence in their files. They made the photographic
36

evidence from five of their photographers, Joe McAulay, Harry Cabluck, Jerrold Cabluck, George Smith
and William Davis, available to everyone. Even
though the Telegram's editorial stance was eventually pro-Warren Commission, anti-conspiracy, the
photographers, editors and the lady who ran the
photo files were all very cooperative.
George Smith's photos showed the three members of
the assassination team under arrest. Jerrold CabIluck's aerial photos became very useful inestablishling Dealey Plaza landmarks and topography. Joe Mc\Aulay's photos of a man arrested in Ft. Worth in
c~~~ection with the shooting may yet become valuable.
TV Station WFAA

The second shining light in Texas was TV station
WFAA, an ABC affiliate. WFAA was very cooperative
(albeit expensive) in providing copies of all their
photographic evidence. TV sequences by Tom Alyea,
Malcolm Couch, A. J. L'Hoste and Ron Reiland, were
made easily viewable and copies available. Much of
this evidence, demonstrating conspiracy, was also
sold to TV networks and newsreel companies.
WBAP -

Ft. Worth

The NBC affiliate in Ft. Worth, WBAP, was less
cooperative. Even though public statements were
made that viewing of Dan Owens and Jim Darnell's
footage was possible, many roadblocks were thrown
into the path of researchers. As mentioned in the
section on NBC, Darnell 'sfootage of the knoll and
parking lot, in particular, is very important. Yet
it has remained effectively unavailable at WBAP.
KTTV - Dallas

Independent TV station KTTV in Dallas has also
suppressed, or at least lost, valuable evidence of
conspiracy. Don Cook's TV footage contains twelve
important sequences, known to the author by interviewing Cook, and seeing him in various other photos
taking his footage. One in particular is a sequence
of a man being arrested in front of the Depository
building at about 1:00 p.m~ From other evidence it
is possible to determine that the man may be William
Sharp, an important participant in the assassination. Cook can be seen in a picture taken by Phil
Willis pointing his 16 mm TV film camera directly at
the man from about ten feet away.
Willis' photo does not show the man's face, only
the policemen arresting him, and his arms and shoulders. Cook's close-up footage, then, is very important. In 1967, the author interviewed Cook in Dallas and determined that his film had been turned
over to the editor at KTTV. A phone call to the
station resulted in a statement being made to the
author that Cook's footage had been lost, "on the
cutting room floor," and was not available for viewirtg. No further efforts have even been made to open
up KTTV's evidence in the assassination.
New Orleans Newspapers

As unbelievable as it may sound to the reader
not living in New Orleans, the only two publications
in the United States printing the truth about the
trial of Clay Shaw were the New Orleans Times Picayune and the New Orleans Times Herald. All other
newspapers and magazines printed completely distorted and untrue versions of the trial. 6
6. For a comparison of New Orleans newspapers and
all other media coverage of the Shaw trial, see the
author's unpubli shed book The Tri al of Clay Shaw The Truth and the Fiction.
COMPUTERS and AUTOMATION for July. 1973

Between 1963 and 1967, both New Orleans newspapers used AP and UP stories on most of their coverage of the Kennedy assassination. Suddenly, because
of the nosiness of two Times Herald reporters, the
pap~rs fourid themselves deeply involved in the middle of the sensational Garrison investigation, and
finally in 1969, they reported on the Shaw trial.
The papers took no editorial position on Jim Garrison, the trial, the investigation, the assassiriation, or the guilt or innocence of Shaw until after
the final verdict was delivered by the jury. Then,
both papers savagely attacked Garrison on the editorial page, joining all other publications in doing
so. Off the record, the reporters and others at
both papers supported Garrison. This was reflected
in a book published by the two Herald reporters,
Rosemary James and Jack Wardlaw, Plot or Politics,
The management and editors of the newspapers evidently paid more attention to national forces from
Washington and New York than they did to New Orleans
citizens or the testimony at the trial. Today,
1973, interviews with 9 people out of 10 in New Orleans will produce the opinion that Garrison was
right, was not and is not crazy and crooked, and
that there was a conspiracy.
The reason for this seeming anomaly is that most
New Orleans citizens read either the Herald or the
Picayune. The verbatim proceedings at the Shaw
trial, as well as all of the detailed events for the
two years that the Federal Government successfully
delayed the trial, were faithfully printed in both
papers. You and I, dear reader, were treated to a
highly biased account of the three years of events
in New Orl~ans by Time magazine, Newsweek, U.S.
News, The New Yo rk-rImes, NBC, CBS, ABC, UP, AP and
all the res t.
For example, the average New Orleans citizen was
well aware that the Justice Department, under both
Ramsey Clark and John Mitchell, was responsible for
continually delaying the trial. You and I were fed
the impression that Garrison delayed the trial.
Mr, New Orleans citizen, lets call him Joe, knew
that Shaw's lawyers were paid by the CIA. You and I
were told that Shaw paid a lot of money for his lawyers and suffered financially because of it.
Joe knew that the FBI was looking for Shaw under
his alias, Clay Bertrand, before lawyer Dean Andrews
ever mentioned the name as being associated with Lee
Harvey Oswald just before he died. You and I were
told that Andrews made up the name Clay Bertrand out
of whole cloth, and no mention was made to us of the
FBI's search.
Joe knew that twelve people saw Clay Shaw together with Oswald and David Ferrie on many occasions and exchanging money on two occasions. You
and I were led to believe by Time and The New York
Times that only three people saw them together and
that the three were not credible witnesses.
Joe knows today how Garrison is being hounded and
framed by the Justice Department in a fake pin ball
rap. More importantly, the government does not want
Regis Kennedy, FBI agent, and Pierre Finck, Army
doctor at the JFK autopsy, to have to testify again
at a trial. They both took executive privilege and
long distance telephone advice from Mitchell in
Washington during the Shaw trial.
Finck's testimony was printed in the Times Picayune but not in Time magazine. He said that an Army
COMPUTERS and AUTOMATION for July. 1973

general gave orders during the autopsy at Bethesda
Naval Hospital. The general, unidentified, told
Finck and the other doctors not to probe the neck
wound in the President. We did not read about this
or hear about it. We also are not up with Joe in
New Orleans in realizing the importance of probing
(or not probing) the neck wound. If that wound did
not go through the President's neck, then conspiracy
would be proved.
The Times Picayune record of the Shaw trial is
especially accurate, appearing in the morning paper
as an almost verbatim report' of wha~ happened, The
Herald's record was reasonably accurate, but because
the paper was printed by 3:00 p.w., the paper missed
some of the longer sessions.
WDSU-TV - New Orleans

As mentioned in the section on NBC, WDSU became
directly involved in the JFK assassination aftermath
because of Rick Townley and Walter Sheridan, ~oth
are, still under indictment by Garrison for bribing
witnesses and tampering with evidence. Townley, on
the staff of WDSU, was close to the actiori with Garrison, Shaw, Andrews, Ferrie, Perry Russo, Layton
Martens, Gordon Novel, Sergio Arcacha Smith, David
Lewis, David Llewelyn, Guy Bannister, and many other
participants in the drama known to him.
According to accounts in the New Orleans papers
and repeated in Paris Flammonde's book, The Kennedy
Conspiracy, Townley tried to get Perry Russo, Garrison's prime witness at the Shaw trial, to change
his testimony at the upcoming trial to make it seem
that Garrison had hypnotized him and then asked
leading questions to get Russo to testify against
Shaw.
Townley went to Russo's house twice, threatened
him with being discredited, perhaps fired from his
job, and offered him a chance to work closely with
NBC in their efforts to "destroy Garrison and his
case". Townley told Russo he could get Shaw's lawyer, F. Irving Dymond, to go easy on him if he would
alter his testimony. He assured Russo that his employer, Equitable Life, had promised the president
of NBC that no retaliation would be taken against
Russo ifhe cooperated with WDSU and NBC.
Walter Sheridan told Russo that NBC and WDSU
could set him up in California where Russo always
wanted to live, if he did help bust up the Garrison
probe. NBC would pay his expenses there, protect
his job, get Russo a lawyer and guarantee that Garrison would never extradite him back to Louisiana.
Sheridan told Russo that NBC had flown Gordon Novel
out of Louisiana to McLean, Virginia (home of the
CIA) and had given Novel (an important witness for
Garrison's case) a lie detector test. Sheridan said
NBC would make sure Novel would never be extradited
to Louisiana to testify. (Novel never was.)
Townley also tried to influence Marlene Mancuso,
former wife of Gordon Novel, and an important Shaw
trial witness. He told her that she should cooperate with WDSU and NBC because Garrison was going to
be destroyed and that NBC was not merely going to
discredit the probe. He said Garrison would go to
jail.
On July 10, 1967, Richard Townley was arrested
and charged with attempted bribery and intimidation
of one witness, and of attempted intimi~ation of
another. Also he was accused of serving as an intermediary to influence cross-examining trial attor37

neys not to damage the character and reputation of
Perry Russo.
Sheridan was arrested on July 7 on the same first
two counts. Both posted bond, and the trials have
never taken place. However, Townley's statements
did come true~ The Federal Government, aided and
abetted by WDSU and NBC, did crucify Garrison. That
case isn't over yet, though, with F. Lee Bailey representing him and the government's key witness,
Pershing Gervais, turning' "traitor" against the government case.
The author's belief is that this kind of behavior
in the face of all the evidence gathered by the
staffs of their own organizations, on the part of 15
to 24 major-news-media management groups, is highly
unlikely.
One other rather minor possibility has been suggested. It is comparable to the kind of influence
obviously wielded recently in the IT&T-Republican
National Convention case. It might be called business or financial or lobbying influence in reverse.
This theory holds that each major news organization
shut up about the Kennedy assassination because they
were afraid of losing face or influence or FCC licenses or business or advertisers or Government favors of one kind or another.
This theory is perhaps best exemplified by a
story told by Dorothy Kilgallen, before she died, to a
close friend. Dorothy had been writing several articles about the JFK assassination for the newspapers who published her column. She believed strongly
that there had been a conspiracy and that Jack Ruby
was involved. She interviewed Ruby alone in his
jail cell in Dallas, the only person outside of the
police who had this opportunity. She told her
friend shortly afterward that she was planning to
"blow the case wide open" in her column. She said
the owner of the New York newspaper where her column appeared was refusing to let her print stories
in opposition to the Warren Commission. When the
friend asked her why, Dorothy said, "He's afraid
he won't be invited to White House parties any
more".
Of the three possible motives for suppression in
the news media, the influence from the top and from
high government places seems the most probable. When
will we, as Americans, learn the truth about this
influence in the case of the Kennedy assassination?
Conclusions

What conclusions can be reached from the information presented herein? Certainly, the pattern of
internal knowledge of conspiracy followed by complete suppression of such knowledge is too strong to
ignore. Two major alternative conclusions suggest
themselves as one reviews the evidence of suppression and secrecy. The first is that our national
news media were controlled On the subject of the assassination~some very high level group in Washington. The control was exercised from the top
down. The top may not have been the very top (chairman of the board), but was near the top. For example, someone in Washington could have told Mr.
Salant at CBS and Mr. Bancroft at The New York Times
what to do, without contacting Mr. Paley or Mr. Ochs.
The orders to cease, desist, and suppress ~
from the top down in each case. To influence the
very top level of all fifteen major news media organizations in this manner would have taken a great
deal more than money or power or threats. In fact,
38

the only kind of appeal which seems likely to have
had a chance of shutting everyone up is a "highly
patriotic, American flag, national security, danger
from exposing the truth" kind of appeal. It is
probably just such an argument that worked with the
Warren Commission. Judging by the fact that Lyndon
B. Johnson said there was a conspiracy to Walter
Cronkite and then successfully persuaded CBS to edit
out his remarks "on ground of national security,"
this kind of an appeal does work. In that case,
enough time had passed that someone at CBS felt
guilty enough about it to leak the information.
The other major possibility seems rather remote
from a probability standpoint, but nevertheless
should be considered. It is that all fifteen (or
really 24) news organizations reached a point of exasperation, disbelief and tuning-out type psychological effect in 1968-1969. In this theory, surprisi ngly bel i eved by a nur,lber 'of peopl e who were i nvolved, the top manager~ of these 24 organizations
reached this exasperation point independently of one
another. Within a two to three-year period culminating in the Shaw trial and a fadeout of belief in
Jim Garrison personally, everyone of these managers
finally said, "Stop, cease, desist, lock the files,
you're fired, shut up, I don't want to hear another
word about it".
0

Unsettling, Disturbing, Critical
Computers and Automation, established 1951 and
therefore the oldest magazine inthe field of computers and data processing, believes that the profession of information engineer includes not only
competence in handling information using computers
and other means, but also a broad responsibility,
in a professional and engineering sense, for:
-- The reliability and social significance of
pertinent input data;
-- The social value and truth of the output
results.
In the same way, a bridge engineer takes a professional responsibility for the reliability and
significarice of the data he uses, and the safety
and efficiency of the bridge he builds, for human
beings to risk their.lives on.
Accordingly, Computers and Automation publishes
from time to time articles and other information
related to socially useful input and output of data
systems in a broad sense. To this end we seek to
publish what is unsettling, disturbing, critical
-- but productive of thought and an improved and
safer "house" for all humanity, an earth in which
our children and later generations may have a future, instead of facing extinction.
The professional information engineer needs to
relate his engineering to the most important and
most serious problems in the world today: war,
nuclear weapons, pollution, the population explosion, and many more.

COMPUTERS and AUTOMATION for July. 1973

THE FLY, THE SPIDER, AND THE HORNET
by Edmund C. Berkeley
Once a Fly, a Spider, and a Hornet were trapped inside a
window screen in an attic. For several hours they walked up
and down, left and right, here and there, all' over the screen.
They could look through the screen at the summer woods,
feel the summer breezes, and smell the summer smells; but
they could not find any hole to pass through the screen to
the woods and fields, so tantalizingly close, yet so far away.
Finally, they decided to hold a conference on the problem of getting through the screen.
The Fly spoke first and said, "My Colleagues, I have surveyed this screen for many hours without finding any hole.
But I cannot believe that there is no hole in this screen. All
my experience of the world up to the present time has shown
me that there are holes in thickets of twigs, holes in screens
of leaves, holes in tangles of grass, holes everywhere. I n fact,
I have been able to fly over, under, or around every barrier I
have ever encountered. Nature does not make thickets,
screens, and tangles without holes.
"Therefore, the principle to be used is perseverance in
spite of obstacles. I will fly again and again and again at the
screen in hopes of getting through. As the old saying goes,
'If at first you don't succeed, try and try again.' I'll never
give up; my honor as a Fly is at stake."
The Spider spoke second and said: "My colleagues, I too
have surveyed this screen for many hours without finding
any hole. I am sorry I have to contradict my honorable Colleague the Fly, but I have come to the conclusion that this
screen has no hole.
"Therefore, the principle to be used is adjustment to
changed conditions. So I will plan to spend the rest of my
life inside this screen. I shall build my web inside the screen,
catch my food, and live out my life in this new way."
The Hornet spoke third, and said, "My Colleagues, I too
have persistently surveyed this screen for many hours with-

out finding any hole. But I remember that I flew into this
general region without going through that screen. Accordingly, there must be some other way of returning to the
woods and the fields.
"Therefore, the principle to be used is exploration of
other alternatives. I shall stop my fruitless search for a hole
in this screen, and instead search in other directions and
avenues to find other means of escape."
The Fly and the Spider said at once, "Honorable Colleague, please pardon us for saying so, but that is plain silly.
Just think how close you are to escaping through the screen
- only a few little thin wires between you and freedom."
The Hornet replied, "My friends, those wires may be few,
and little, and thin, but nevertheless they are too substantial
for me to pass through."
The Fly then said, "Besides, honorable Colleague, this
conference of ours was called on the subject of getting
through the screen - and, honorable Colleague, you are off
the subject and out of order." The Spider said, "I must
agree with the Fly."
Thereupon, the Hornet made a motion that the subject of
the conference be amended from getting through the screen
to escaping to the woods and fields. But he was outvoted 2
to 1.
None of the three would change his views, and the conference soon adjourned.
A day later, the Fly died of exhaustion.
A week later, the Spider, having caught only one moth in
his web, died of starvation.
But an hour later, the Hornet exploring under the eaves
of the roof between the inclined beams of the attic, found
an opening to the outdoors, and flew out, returning to a
happy life in the woods and fields.

Taken from:

"RIDE THE EAST WIND: Parables of Yesterday and Today"
by Edmund C.

Berk~/ey,

Over fifty parables (including allegories, anecdotes and
fables) by Berkeley and 20 other authors, modern and ancient, dealing with famous problems, modern, classic, or ageless. Each decorated with a bouquet of proverbs and quotations for readers who like to choose which variety of lesson appeals to them. An appetizing guide to many topics of
common sense and wisdom. Illustrated, hard cover, over
250 pages, to be published by Quadrangle Books, a New
York Times Company, in October. An ideal gift.
ORDER YOUR COpy NOW
Returnable in 7 days for full refund if not satisfactory.
HOW CAN YOU LOSE?
COMPUTERS and AUTOMATION for July. 1973

Author and Anthologist

- - - - - - - (may be copied on any piece of paper)- - - - - - - - - - - -

To:

Computers and Automation
815 Washington St., Newtonville, Mass. 02160
Please send me when published (October publication
expected)
copy(ies) of Ride the East Wind:

Parables of Yesterday and Today by Edmund C.
Berkeley, Author and Anthologist. I enclose $7.25.
($6.95 price + 30<;: postage and handling per copy)
Total enclosed

(Prepayment is necessary)

Returnable in 7 days for full refund if not satisfactory.
My name and address are attached.
39

ACROSS THE EDITOR'S DESK
Computing and Data Processing Newsletter
Table of Contents

APPLICATIONS

From Shoes to Bumpers - Computers Help
En sure Qua l i ty
Binder + IBM Computer: Covering a Mile
of Magazines
Minicomputer Helps Bay Area TV Station
Boost Ratings During Election Coverage
Yogi Bear and Computer Hard At Work at
FUn Center
Saturn Computers Wrestle with Exploration,
Pollution and Flooding
Iran's Computerized Monitoring and
Control System Regulates Crude Oil
Pipeline Flow

40

42
42

40
EDUCATION NEWS

41
41
41

42

APPLICATIONS
FROM SHOES TO BUMPERS COMPUTERS HELP ENSURE QUALITY
George Snyder
Mobay Chemical Company
Div. of Baychem Corp.
Pittsburgh, Pa. 15205

Two new computers have been installed at Mobay's
headquarters here to provide management with the
most advanced tools for solving both business and
sCIentific problems. The IBM computers are used for
such diverse challenges as providing an extra step
from the soles of your shoes and a tougher, more
protective bumper on your car. Mobay sells thousands of tons of basic chemicals and plastic resins
each year to a variety of customers, including companies in the automotive, clothing, construction,
furniture, electronics, aircraft and appliance industries.
A small computer, the IBM System/7, is attached
to a testing machine which presses, pushes and pulls
flexible and rigid foam and plastics samples into
various shapes to test resilience and strength. The
computer records and prints out periodic readings as
the material is put through its contortions. This
pretesting enables Mobay to assure its industrial
customers that the chemicals they purchase, when
properly formulated, will make strong, quality shoe
soles, automobile bumper components, or numerous
other consumer products.
The second larger computer, an IBM System/370
Model 135, is used for a variety of control operations, from keeping an eye on construction costs to
the precise color matching of plastics. Contained
in the larger computer's information storage units
is color spectrum data which directs the mixture of
pigments to achieve uniform colors in the pelletized
plastics produced by Mobay.
Mobay customers mold this material into such
products as football helmets, casings for pens, and
40

Computer to Help Design "Eyes" for
Space Probes
Air France Uses Minicomputer to Assign
Seats in Airplanes; Saves $35,000

United Nations International School Gets
Free Computer System
American College Abroad Adds
Computer Sciences

43
43

COMPUTER RELATED SERVICES

Worldwide Cancer Information Service Center

43

even household items such as chairs and tables.
Called MerIon polycarbonate, it also is made into
clear plastic sheeting which can be used as window
material in areas where vandalism and breakage are
serious problems. In laboratory tests, the material
resisted .38 cal. slugs, shotgun blasts and thrown
bricks.
Mobay's plants in West Virginia and Texas are
tied into the main System/370 in Pittsburgh via
telephone lines and their own small, general purpose
computing systems, the IBM 1130. This teleprocessing system is used for inventory reporting, payroll
processing and for maintaining all plant records
stored in the larger computer in Pittsburgh. The
System/370 also is used to monitor both minor repair
and construction costs and as an aid in plant design
calculations. It performs major business tasks in
marketing and financial areas, and is programmed to
handle Mobay's accounting activities.
BINDERY + IBM COMPUTER:
COVERING A MILE OF MAGAZINES
John Pilkington
The Heckman Bindery, Inc.
North Manchester, Ind. 46962

If Heckman Bindery could stack all the magazines
it handles each year into a single pile, the pile
would rise well over a mile into the air. Heckman
puts hard covers on these periodicals for libraries.
The northeastern Indiana firm also rebinds old
library books, for a potential savings of $50 over
the cost of new books; prebinds new volumes to add
months to their usability; repairs and restores Bibles and other valuable books; and binds inexpensive
paperbacks so they can be circulated in hard cover
form for as long as two years.
Just keeping track of the libraries' binding instructions for their periodicals was a monumental
task; 250,000 different work order forms were on
file. Librarians complained because it took so long
to dig through their files for specifications and to
fill out the detailed forms Heckman needed.
COMPUTERS and AUTOMATION for July. 1973

Both problems were solved with a computer, an IBM
System/3, and a new order-handling procedure called
H.E.L.P. (tleckman ~lectronic ~ibrary frogram). The
firm claims its program can save librarians up to 50
percent of the time it normally takes to maintain
and prepare their periodical binding specifications.
As an example of how H.E.L.P. works, all of Purdue University's constant binding instructions for
TIME magazine are stored in the computer. Heckman
has supplied the University with a computer-printed,
mUltiple-page form, called a "pattern", and an index
card containing this information. If a Purdue librarian wants to submit the school's January through
September, 1972 issues of the magazine for binding,
he simply pulls out the TIME pattern, enters the
variable information (dates and volume number), and
sends the completed pattern to the bindery with his
magazines. Heckman uses the pattern to direct its
work and update its computer records, if necessary.
The index card is retained in Purdue's file to
alert librarians that the January-September issues
of TIME are at the bindery. The back of the card,
which contains information on each binding order,
serves as a chron.ological history for the University's permanent files. When the bound volume is
delivered to Purdue by a Heckman representative (the
company provides 28-day pick-up and delivery service
in 33 states), it is accompanied by a new pattern.
This enables the University to repeat the ordering
process in the following quarter.

YOGI BEAR AND COMPUTER
HARD AT WORK AT FUN CENTER
Thomas J. Olds
Kings Island
P.O. Box 400
Kings Mills, Ohio 45034

Though these furry cartoon characters seem a bit
fuzzy about the goings on of a computer room, there's
no misunderstanding over their common goal of helping people have more fun at Kings Island. A 1,600acre family entertainment complex north of Cincinnati, Ohio, Kings Island reopened April 28 for its
second season. With a 33-story replica of the Eiffel Tower as its landmark and symbol, the fun center
features more than 100 attractions in addition to a
motor inn, two golf courses and 300 campsites. More
than two million people from alISO states visited
the facility during 1972.
Drooper and Bingo of the Banana Splits (who seem
to be having trouble with the computer printout),
along with Yogi Bear, Fred Flintstone and other cartoon characters, perform in the Happy Land of HannaBarbera, one of five themed areas.

Another feature of the H.E.L.P. system is the assistance it provides both Heckman and the libraries
in purging their files.
MINICOMPUTER HELPS BAY AREA TV STATION
BOOST RATINGS DURING ELECTION COVERAGE
Edgar E. Geithner
Data General Corp.
Southboro, Mass. 01772

A Nova-based election reporting system gave a San
Francisco, Calif. television station the greatest
election viewer ratings in its history during the
1972 elections. The Nova 1200, made by Data General
Corporation, is part of a system that receives, tabulates, stores, and displays voting information on
command, giving viewers the latest tallies and percentages on their screens. A Telemation Event Display system designed by Telemation, Inc. of Salt
Lake City, was installed just before the primaries.
During the primary elections, California had an
unusually large line-up of names and issues. Mr.
Briggs said the old method of typing lists on paper
took three times more people and ten times longer to
prepare for a newscast. The Nova-based system updated the entire vote tallies in just three minutes.

The computer, an IBM System/3, operated here by
Lois Anson, helps monitor by state and county a
sampling of auto license plates to determine which
areas should be stressed in Kings Island's promotional efforts. The IBM computer also is used for
gift shop inventory and retail sales analysis reports, which help managers stock their shelves according to the popularity of the various items in
their shops.

Election information :nformation was prepared in
advance on a character generator keyboard in "pages".
A page lists district, contest, candidates and party
affiliation. It is numbered for recall and updating.
Shortly before KRON-TV's election reporting began, a
program tape was generated from the pages in disc
storage and loaded into the Nova. The pages were
displayed on a terminal and televised.

RCA Corp.
30 Rockefeller Plaza
New York, N. Y. 10020

After the polls closed, reporters called vote
totals into the news desk. Six teletypewriters were
used to input vote tallies to the Nova. The viewer
ratings justified the expense of the Nova-based system according to Clark Briggs, KRON-TV's election
producer.

The launching of the first Skylab manned space
station and the study of flood control on Alabama
rivers have one thing in common: they're both relying on a family of computers known as the RCA 1l0A' s.
Originally developed for NASA, the computers were
installed at Cape Kennedy to check out the Saturn

COMPUTERS and AUTOMATION for July, 1973

SATURN COMPUTERS WRESTLE WITH
EXPLORATION, POLLUTION AND FLOODING

41

IB and Saturn V rocket boosters up to the instant
of launch, and at the Marshall Space Flight Center,
Huntsville, Ala., for use in developing and testing
the launch vehicles prior to delivery to the Kennedy
Space Center.
One of the RCA 110A computers recently was loaned
by the Marhsall Space Flight Center to the University of Alabama's electrical engineering department
in Tuscaloosa, where it is concerning itself with
such problems as flood control on Alabama rivers,
sewerage disposal problems in Mobile, and working
out complex problems for graduate and undergraduate
students.
While the academic 110A began playing the quiet
role of electronics professor, two of its sister
computers at Cape Kennedy were acting more like
robot physicians as they put the Saturn V and Saturn
IB rockets through a thorough physical examination
prior to launch. The goal of Skylab is to conduct
scientific missions that could lead to better utilization of the earth's resources and to test man's
ability to live in space for long periods.
In Tuscaloosa, things may be less exciting for
the ac~demic 110A, but they certainly are no less
important. Environmental research is a high priority item and this Space Age computer may help man
keep his feet on the ground.
IRAN'S COMPUTERIZED MONITORING AND CONTROL
SYSTEM REGULATES CRUDE OIL PIPELINE FLOW

J. N. Dravillas
Westinghouse Electric Corp.
Gateway Center
Pittsburgh, Pa. 15222

Iran's largest completely computerized pipeline
supervisory control system will be installed on the
new Ahwaz-Rey Second Crude Oil pipeline in August,
1973. The monitoring and control system, perhaps
the most modern in the industry worldwide, will provide centralized control of 466 miles (750 kilometers) of crude oil pipeline from Ahwaz~ in the south
of Iran, over the Zagros Mountain Range, to the Rey
Refinery, just south of Tehran. The pipe diameter
is 30 inches and 26 inches, and the line will have
a future capacity of 410,000 barrels a day.
The control system includes twin Westinghouse2500 computers (one serves as a backup) at the central control facility at Rey and control and monitoring equipment at seven remote stations along the
line (with two more planned for the future). Because
of the range of desert and mountain temperatures to
which the system will be subjected, equipment at the
remote stations is designed to operate satisfactorily
from -22 to +167 degrees F (-30 to +75 C). At t~e
central station at Rey a graphic representation
clearly illustrates the up-to-the-minute situation
along the entire length of the line -- the central
computer scans all remote instrumentation every two
secondsl
Since this is a supervisory control system, the
computer takes no direct action to rectify any problems or vary the flow rates. Instead, it feeds pertinent information to the operator on hand, who then
takes corrective action.
The Ahwz-Rey pipeline is owned by the National
Iranian Oil Company; general contractor for the job
is Neill-Price International.

42

COMPUTER TO HELP DESIGN
"EYES" FOR SPACE PROBES

Itek Corporation
Lexington, Mass. 02173

A "virtual storage" computer has gone into operation at the Optical Systems Division of Itek Corp.
An IBM System/370 Model 158 will perform the millions of calculations needed to design and produce
lenses for cameras on space ships, astronomical
telescopes and other special devices. The computer
will serve Itek's divisions, processing business as
well as scientific and engineering applications and
will also become the base of a corporate management
information system.
Virtual storage, as incorporated in the Model
158, is a technique for storing data that makes the
computer appear to the user as if it had a capacity
of 16-million bytes of storage -- many timesi ts actual capacity. The technique can greatly speed and
simplify the development of many computer applications, including remote ~mputing networks and systems for inquiring into the computer from distant
locations.
Itek, which has run computer programs on a Model
158 at speeds up to five times those possible with
their installed System/370 Model 145, relies heavily
on computers for lens design and production. This
is an extremely complex and demanding art, according
to Dr. Robert Hills, Director of Optics. "Today,
some lens design projects would take years without a
computer. The computer can also accurately predict
the performance of lens systems before manufacturing
begins. It would be prohibitively costly, if not
impossible, to carry out dependable predictions by
any other means."
One Itek-developed fabrication process requIrIng
a powerful computer is CAOS -- Computer Assisted Optical Surfacing. A large optical surface can be
ground and polished by a special machine which is
controlled by computer-generated instructions. These
instructions are produced by the computer from optical test data. Other technical applications will
include design and simulation of circuits and thermal, electrical and structural devices.
AIR FRANCE USES MINICOMPUTER TO ASSIGN
SEATS IN AIRPLANES; SAVES $35,000

June Fanazick
Public Relations
Data General Corp.
Southboro, Mass. 01172

A minicomputer is saving Air France approximately
$35,000 annually according to a spokesman for the
airline. The computer, a Data General Corporation
Nova, reduced the time needed for passenger checkin at John F. Kennedy Airport, New York, the Air
France representative said. The Nova 800 minicomputer is part of the Multiflight Boarding Pass System developed by General Computing Equipment Corp.
of Garland, Texas, which was specifically designed
to meet passenger check-in time demands caused by
wide-bodied jet transports.
With the help of the Nova, passenger check-in
times are significantly reduced, and over-boarding
and redundant seat assignments are virtually eliminated. Each agent has a display console with an array of lights that show specific seat status, al-

COMPUTERS and AUTOMATION for July. 1973

lowing him to assign seats and issue boarding passes
automatically. Both passenger's and stewardesses'
portions of the pass contain flight number, class,
seat number, destination, date and boarding gate.
The entire check-in process takes as little as 15
seconds.
When an agent's display shows that a desired seat
is available, he reserves that position by pressing
the appropriate seat button. After the boarding
pass is printed, the system transmits the transaction to the Nova's memory, removing the seat from
the pool of unoccupied seats and updating all other
agents' displays. It assigns group seating in a
s imil ar manner.
The system adapts to a variety of seating configurations, such as four, five or six across, for different flights and different flight segments within
a flight. Red and blue lights indicate cabin class,
exits and nonsmoking zones. The Nova stores these
and any other special programmed cabin designations.
A single central control station operates up to
15 consoles; system expands in additional increments
of 15 consoles. A key lock controls access to the
system for security measures. In the event of power
failure, the system retains all registered seat assignments until power is restored.
Air France at JFK also services Avianca and
Transportes Aereos Portugueses. A similar system
was installed' at Philadelphia International Airport in March that services Air France, Alitalia and
Lufthansa Airlines. Air France also has systems at
Los Angeles and Montreal.

EDUCATION NEWS
UNITED NATIONS INTERNATIONAL SCHOOL
GETS FREE COMPUTER SYSTEM

Mrs. Murray Fuhrman
Special Representative of the Secretary-General
United Nations
New York, N. Y. 10017

The United Nations International School has announced the gift of an educational computer system
from Digital Equipment Corporation's Educational
Products Group in Maynard, Mass. The $33,500, "EduSystem-25", a powerful, mul ti-functional minicomputer
system, has been donated with free teacher instruction, a comprehensive kit of textbooks, teacher
guides, visual aides, and sample programs. The system will be shared by students from more than 90
countries at the new United Nations International
School building on East River Drive at 25th Street.
The Digital system is designed primarily for instruction in mathematics problem solving, beginning
at grade seven. It also runs simulations in physics, chemistry, biology, earth sciences, and social
studies. Additionally, the equipment c~n be used
for remedial and tutorial drill and practice in
mathematics and in a variety of other subject areas.
Upon receipt of the computer system, th~ Secretary-General of the United Nations, Kurt Waldheim,
said "It is quite possible that the students from
throughout the world who face each other across the
Digital computer may, in a few years time, face each
other across international conference tables. The
lesson of using teamwork to achieve common goals
COMPUTERS and AUTOMATION for July, 1973

will provide the foundation upon which a significant
contribution to international cooperation may be
made tomorrow."
AMERICAN COLLEGE ABROAD ADDS
COMPUTER SCI ENCES

Public Relations Department
National Cash Register Co.
Dayton, Ohio 45409

Computer science instruction has come to the American College of Switzerland. Students at this unique
American institution in Europe now can study data
processing just as their counterparts in the United
States. As a part of this training the American
College is offering intrOductory courses in computing, courses in COBOL and FORTRAN, and two advanced
data processing seminars in its International Business Studies and Natural Sciences departments.
To accomplish this the college has installed an
NCR Century 100 system with a 32K memory, 450-lineper minute alphanumeric printer; 300-card-per-minute
reader, and a dual-spindle disk unit which can store
9.4 million bytes.
The expansion of American business activities
overseas, the development of the Common Market and
the rapid multiplication of U.S. missions throughout
the world have combined to create a new class of
Americans and a new problem. Aside from the difficulties and expense involved in sending young men
and women back to the United States for their college education, "many U.S. families living overseas
prefer to have them study in a European setting,
both for the cultural advantages and because of
family proximity.
While designed for the U.S. student abroad, the
American College is not restricted to this group.
Stateside students from college? which do not offer
a junior-year-abroad program are also enrolling in
the American College of Switzerland in ever-growing
numbers. The College also accepts qualified nonAmerican students who wish to obtain an American undergraduate education or to prepare for graduate
studies in their chosen field in America.

COMPUTER-RELATED SERVICES
WORLDWIDE CANCER INFORMATION SERVICE CENTER

James Fordham
National Cancer Institute
Bethesda, Md. 20014

The U.S. Department of Health, Education, and
Welfare/National Institute of Health has announced
the launching of the first worldwide service to
speed vital cancer information to physicians and
scientists. Dr. Frank J. Rauscher, Jr., Director of
the National Cancer Institute, said NCI has awarded
a $780,000 contract to Informatics Inc. of Canoga
Park, California, to develop the first phase of the
Cancer Information Service (CIS).
The IS-month contract will include the most extensive study in history of worldwide cancer data
services, the information needs of cancer scientists, and how information about cancer can be made
more available throughout the scientific community.
(please turn to page 49)
43

NEW CONtRACTs
'TO

GTE iHf8rfiliiH8H Sy§iems file: I
White pi~iHSI

N;y:

~66i~i ~gttltit~ AdminiiiratiOHf"tl~Pt: of H~W, W~§hih§=

toR I oJ;:

Computer Sciences corp.-tEse);Los Angeles, Calif;

Goodyear Aerospace Corp.,
Akron, Ohio
Univac Div., Sperry Rand
Corp., Blue Bell, Pa.
Librascope Div., Singer Co.,
Glendale, Calif.

Australian Mutual Provident
Society (AMP) , Sydney, Australia
Pasadena Laboratory, Naval
Undersea Center, San Diego,
Calif.

ifi6:i Ft~hkliH Lak~s, N.J.

InstrumentatiOn Engineering,

Siemens AG, Munich, West
Germany

Space & Defertse Systems Div.,
Fairchiid Camera & Instrument
Corp., Syosset, N.Y.

Naval Electronics Systems
Command, Washington, D.C.

College of Engineering,
University of Missouri-Columbia; COlumbia, Mo.

ty-

of Rochester
Medical Center, Rochester,

Naval ,Electroni-cs" ;);~~~ms
Command, Washington, D.C.

National Cancer Institute

N.Y,

Loglevn, Inc.,
Calir.

Torranc~.

Experiment~I

eeRfH H~AFBG JI

FAA, Atlantic Citi, ~;J:

§t8Hr6fJK~§~oreh Institute
Menia ~afk, Colit,

National Science Foundation

Logicon, Inc., Torrance,
Calif.

Fort Monmouth, N.J.

Informitle~ tHe:; W~st~~n
Div., Canoga ~af~t Galif;

Ames Research Center, NASA,
Mountain View, Calif.

Advanced Memory Systems, Inc.,
(AMS), Sunnyvale, Calif.

TeletunkenComputer Gmbh,
Konstani, West Germany

Collins Radio Co., Dallas,
Texas

Pan American World Airways

Computer Corporation of
America, Cambridge, Mass.

Dept. of Defense, Washington,
D.C.

Inforex, InC., Burlington,
Mass.

Social Security. Administration, Federal Business Office,
Washington, D.C.

Communication Systems .Div.,
Philco-Ford Corp., Willow
Grove, Pa.

U.S. Postal Service

Missile .Systems Div. (MSD),
Rockwell International Corp.,
Anaheim, Calif.

Northrop Corp., Columbus,
Ohio

44

Development and inst~llation of nationwide
~~£~ riom~unicationsnetwork to sp~ed supplefuen~~ise~Ufity iHcbme paymeRis is ReeUy
aged and Handicapped persons
Providing comprehensive scientific analy.~i s and programming services to Goddard
S'
£:"ii ht Center GreeHBei-t.;. M{1.;---_ '_.'
Production of Minuteman Wea~bris S ~teill
Controllers (WSC)
Printer order to be delivered over the
next 12 month's

ace

Univac Div;, Sperry Rand
Cbr. BlueBell Pa.
Centronics D~ta Computer
Hudson, N.H.

O~l.V~rs-i

=

AMiIDrff

A~myEr;e[rlHi16'

COllVlland,

$21

miiiiSH
(appro'xliiiat8 J

$16.5 million

Developing and butldiQg f'rst protqtype,
flight sim~lator tor Air Force F-15 Eaglei
McDonn~ll Douglas is prime contractor for
new air superiority aircraft
UNIVAC 1110 system to be center of d~ta
communications network for AMP coveaRff
Australia and New Zealand
Four weapon control and setting subsyste~s
for UnderWater Fire Control System Mk 116
Mod 1 to be instolled' aboard DLGN 38,. 39
and 40 olass vessels
A System 390 computer-controlled test stati6n; will be used tor te~ting telephorie
swi tehllllL lo.gJt} 1.n' W~'Lt Germany

5~cond pno§~ et ~har~~~coupled device (CCO)
tl~V~lbpffi@Ht p.6grgm"wh1ah ~~phfigizeg devel6~ffient bf lafgeraftaY§ to be used in tiva-

rf{l~~ 11~, tIWI::H9I1t::i~~~iimagitlg, applicatIons wItk furtnuf @ffori on anti~blo6ming
and new on-chi am iifi ef t~·
_.
Devel~pment of control system that w11
m~kB §aR~Uine's transmitting sites almost
immune either to BHfJtny a't tack or equipment
failure; Sanguine Commurif~8th'ns System
provides worldwide communications WIth submarines operating at any depth, anywhere
Developing automated pre screening system
for detection of cancer cells; currently
no such system' exists
A digital air traffic control pilot display system to be used to intrease the
flexibility and ~hroughput of NAFEC's
fHyHill JU/lIylatfQn Faci 1 i ty
One yeaff~§~arch itl industrial applicationof advanced aut(1m~ti(jn; exploratory
project desigried to develop techniques and
systems fo~automating materials handling,
inspection and assembly line production
Systems engineering, services in support of
Joint Ground ~rid Amphibious Military Operatiang (BAMO) program
Computer programs for a central computer
and its subsystems, includingmini-computers, to control data acquisition and recording at the wind tunnel test facilities
Design and manufacture of complete memory
sysiems(average size 128K words X 59 bits)
to he s~pplied over a two ye~r period
Airline communications system capable of
handling 52,000 "conversations" an hdur
wit~ reservations computer and can t~ansmit
11,500 teletype messages an hour; Collins
system will be communications arm of PANAMAC
I~worldwide reservations system
Model 204 On-Line Database Management Software System; syst~m performs database management functions from multiple remote terminals
Seventy 1302 data entry systems, totaling
85i individual keystations; co~tract also
provides government with option to increase
total order for Inforex data entry systems
by:approximately 50%
Modifying OCR I electronic mail sorting devices currently installed in 12 Post Offices
throughout U.S.; bar-code reading capabiliti~s are being added
Two air-to-air ranging radars for Northrop's
lightweight fighter

A
[J

I

r
(

it1H

(

[
(
(

]

$4 million'
(appl'()ximate)
$2.6' mii1i8fl
(approximate)
$855,000
$800,000+

$402,220

$347,583
$330,734

$277,100

$192,000$133,000

COMPUTERS and AUTOMATION for July. 1973

NEW iNSTALLATIONs
OF
n·,".

Burroughs B 6700 system

]

Data Cyber 70, Model 73

Control Data
Cyber 70, Model 73

.)~t@ffi

Cyber 70, Mod"el liL_50yH~ffi

Conttol :09H\ 5c-1700 system
Honeywell dual 6050 system

NCR Century 200 system

NCR Century 251 system
NCR Century 300 system

Univac 1106 system

Univac 1108 system
Univac 1110 system

Univac 9400 system

Univnc 9700 system

Univnc 494 system

,..i

11('

.'

.

..

1] t 'r""

and processing 10.5 million expendIture
warr~nts (checks), and to process over 4 ~~!~~onp_
inc.ome, sale.s an.d o.. t~.er tax r. eturns eahh year: re"lac~~ t:Uffe~t Dijt~ ~gHtH gqufpilHWf.
fFst~ms combined valU~·' - 4:6 miHIoH.··· .
Computing and communicatio~s se1;vicesiq Crown ColComputer Data (H.K.) Limiti
o~y customers; special application~, ~nclude off(COL), Hong Kong
ine
anrl
.. on. ~l.·1.·ine
bahkiH.~:
. §tock
§H9rei teg~st~a] un
'fr11
..
("'~l( 1 nc",~r 1
;'r,/-l ( ~...
.
:p
•
-.--. ----'lon' :Structural eni eer1H ,a-rH:l·8- f -track- b-etbn
Providin facul tv memberS 11 '0. s
H s iIi' e
ortunity tdus~ 2smBUIIHH tG8}alaU€~i aHiIE~aHt~ ~!~
colleges within the University will use the system
in a wide variety of applications
Supporting computing requirements of students, facUniversity of Nevada System,
ulty and adrnini stration at campuses in Reno and Las
Reno, Nevada
Vegas, the Desert Hesearch Institute, and the three
communit colle es
Forming the central computer complex of a new comUnion Bank of Switzerland;
mercial bankin'g system; includes dual CD CYBER 1000
Zurich, Switzerlatld
data communications systems; another dual CYBER 1000
system is scheduled to be added in late 1973 in Gen':'
eva, Switzerland! shipm~nt5 include variety of entry/display terminals and peripheral equipment
(s stem valued ~t 12.3 millionJ ...... --- .. Development of regi~nal d~t~ ~~fit~r at the univerfor Culture,
sity to serve a variety of scierttifie research and
education~l institutions in northern West Germany
Supplying customers with expanded data services
Procesa, Zaraooza City, Spain
Use in part of the WOI;ld Wide Mili tar), CommaIJd and
U.S. Army Operations Center~
Cont~o~ SJ-:s,te)11 .(WWMt!j§j i U systeiil~ were del.ivered
Washington, D.C.
iil 1972, remaining 20 systems to be Installed In 1973
(s stem valued ~t 3.2 million)
First phase of nationwide on-line computer network;
also have an NCR Century 100 system at data center
for off-line batch processing; a second Century 200
system will be added later for further expansion of
bank's data processing services·
..
_.
Nucleus of Central Information File systems for
St. Petersburg Bank and Trust Co.,
their member insti tutions
St. Petersburg, Fla.
Nationwide data processing network linking all
La-Z-Boy Chair Co., Monroe,
tfiilnu£'8oturitHJ huil! ties to company headquarters
Mich.
_U.Si.lliti valued at. 797,000)
.
eplacerileht Of ~tfialler system;_.1!J~s ~7.':-~-~~~ taxeolCity of Pittsb~rgh, Pittjburgh~
l'ecH'oill .wat~r\H.~li.~~: ~':71~';a~1 accounting, police
Pa.
it\t~f!9H~l\. ~perations and police traffic tickets
(sis(emvalued at $1,461,000)
Tabulating Consul tants, ·IM.,
Additional processing power and data storage capaBurbank, Calif.
city; will replace two small computers now 1n
(2 s ste'ms
service
Large volumes of work now handled by sta;VlM \\f~~\1s
Conzinc Riotinto of Australia,
(s
stem valued atmor.~:tha"
.
lL. ______ ...,Ltd., Melbourne, Australia
Maryland State Colleges, Towson
Maryland State Colley~$ lnfO'fffiaH'On B'e"t~f
CIG)
State College, Towson, Md.
to serve 7 colleges in th~ §\i~@ i~' luil range of
business/admini%ttatiY~ ~~~lielllons; student use
via tetminals for t"ii!Uhtion in computer language
and operatio", testing various programs, and mathematical ~nd scientific calculations
(s stem valued at a roximatel _ L_
Personnel Research. Div., Human ReResearch and development activltt~i 0
sources Lab., Lackland AFB, Texas
Resources program
City Public Service Board,
Toti:Jl management information system; appJications
San Antonio, Texas
include customer revenue accounti~g, ele~trie and
gas systems ,load information, forecasting COHsump~
tion, resource and facilities management, decision
making model, and coristruction man~~ement system
(system valued at $3.4 million) .
Nova Scotia Power Corp.; Halifax,
Processing approximately 250,000 customer accounts
Nova Scotia
across Nova Scotia; future plans include acquisition of visual display xerminals fg~dealing with
customer inquiries in ,an on-line mode'
.
'. Alfred Herbert, Coventry, England
Centralizing overall ,administration a,ndproduction
f~nctions; c?mpany.p~ans to expand, prpsent p~od"U­
tion schedulIng, dIrect order entry, invento~y Control and sales, and marketing systems into most of
its locations; replaces an IBM System/360-40
'
(system valued at $1 million) .
"
.
Beamtenheimstaettenwerk GmbH
Primary use in administration of various construc(BHW), Hameln, West Germany
tion projects under Association'$ authority; data
communication lines will link o\ier 100 LlHW consulting agencies to data stored in Hameln center

t

system

1·

~~oducing

COMPUTERS and AUTOMATION for July. 1973

-1:

'f

45

CALENDAR OF C'OMING EVENTS
July 17-19, 1973: Summer Computer Simulation Conference,
Oueen Elizabeth Hotel, Montreal, Canada / contact: Stuart
Trask, Sun Life Assurance Co. of Canada, P.O. Box 6075,
Montreal 101, P.O., Canada

Sept. 25-28, 1973: Engineering in the Ocean Environment Conference, Washington Plaza Hotel, Seattle, Wash. / contact: Ted
Hueter, Honeywell Inc., Marine Sys. Ctr., 5303 Shilshole Ave.,
N.W., Seattle, WA 98107

July 20-22, 1973: 1973 International Conference of Computers in
the Humanities, University of Minnesota, Minneapolis, Minn. /
contact: Prof. Jay Leavitt, 114 Main Engineering Bldg., University of Minnesota, Minneapolis, MN 55455

Sept. 25-28, 1973: IFAC Symposium, Purdue Univ., W. Lafayette,
Ind. / contact: Carl Jenks, Div. of Confs., Rm. 116, Stewart
Ctr., Purdue Univ., W. Lafayette, In 47907

July 22-27, 1973: 32nd Research Conference on Instrumentation
Science, Hobart & William Smith Colleges, Geneva, N.Y. / contact: ISA, Meetings Administrator, 400 Stanwix St., Pittsburgh,
PA 15222
July 23-27, 1973: 3rd Annual International Computer Exposition
for Latin America, Maria Isabel-Sheraton Hotel, Mexico City,
Mexico / contact: Seymour A. Robbins and Associates, 273
Merrison St., Box 566, Teaneck, NJ 07666
Aug. 5-8, 1973: 7th Annual Mathematical Programming Seminar
and Meeting, Breakers Hotel, Palm Beach, Fla. / contact: George
M. Lowell, Haverly Systems Inc., 4 Second Ave., Denville, NJ
07834
Aug. 6-8, 1973: 5th Annual Meeting of the National Association
for State Information Systems, Hotel Ambassador, Chicago, 111./
contact: Carl Vorlander, NASIS, Exec. Dir., P.O. Box 5377,
Lexington, KY 40505
Aug. 7-9, 1973: Association for the Development of Computerbased Instructional Systems, University of Michigan, Ann Arbor,
Mich. / contact: G. Ronald Christopher, The Ohio State University, 1080 Carmack Rd., Columbus, OH 43210
Aug. 13-17, 1973: SHARE Meeting, Miami Beach, Fla. / contact:
D. M. Smith, SHARE, Inc., Suite 750, 25 Broadway, New York,
NY 10004
Aug. 20-24, 1973: 3rd International Joint Conference on Artificial
Intelligence, Stanford University, Stanford, Calito/ contact: Dr.
Ma~ B. Clowe;, !..~boratory of Experimental Psychology, University of Sussex, Brighton, gl.i~;g:,! ~N1 90Y, England
Aug. 27-29, 1973: ACM 73, Atlanta, Ga. / contact: Dr. Irwin E.
Perlin, Georgia Institute of Technology, 225 North Ave., N.W.,
Atlanta, GA 30332
Aug. 27-Sept. 1, 1973: Computer Arts Society, 1973 Edinburgh International Festival,., Edinburgh, Scotland / contact: R. John
Lansdown, Secretary, Computer Arts Society, 50-51 Russell
Square, London WC1 B 4JX, England
Aug. 30-Sept. 1, 1973: International Conference on Systems and
Control, PSG College of Technology,Coimbatore, India / contact: Dr. R. Subbayyan, PSG College of Technology, Coimbatore 641004, Tamil Nadu, India
Sept. 4-7, 1973: International Computing Symposium 1973, Davos,
Switzerland / contact: Dr. H. Lipps, I nternational Computing
Symposium 1973, c/o CERN, CH-1211 Geneva 23, Switzerland

Oct. 2-3, 1973: Elettronica 2 - 2nd International Conference of Industrial Applications of Electronics, Turin, Italy / contact: Secretariat, 2nd Covegno Internazionale di Elettronica I ndustriale,
Corso Massimo d'Azeglio 15, 10126 Torino, Italy
Oct. 2-4, 1973: 2nd International Computer-Aided Design and
Computer-Aided Manufacturing Conf., Detroit Hilton Hotel,
Detroit, Mich. / contact: Public Relations Dept., Society of
Manufacturing Engineers, 20501 Ford Re., Dearborn, MI 48128
Oct. 3-4, 1973: 7th Annual Instrumentati~n & Computer Fair,
Sheraton Inn/Washington-Northeast, Washington, D.C. / contact: Richard Bullock, Instrumentation Fair, 10774 Tucker St.,
Beltsville, MD 20201
Oct. 8-12, 1973: Business Equipment Show, Coliseum, New York,
N.Y. / contact: Rudy Lang, Prestige Expositions, Inc., 60 E.
42nd St., New York, NY 10017
Oct. 12-14, 1973: 12th Annual UAIDE Conference, Chase-Park
Plaza Hotel, St. Louis, Mo. I contact: Bobby R. Peoples, HEW,
Rm. 1070 North Bldg., 330 Independence Ave., S.W., Washington, DC 20201
Oct. 15-17, 1973: 14th Annual Switching ,and Automatal Theory
Symposium, University of Iowa, Iowa City, la. / contact: Prof.
Gerard Weeg, Computer Science Dept., Univer.sity of Iowa, Iowa
City, IA 52240
Oct. 15-18, 1973: 28th Instrument Society of America International Conference and Exhibit, Astrohall, Houston, Tex. / contact: Philip N. Meade, Exhibit Director, ISA, 400 Stanwix St.,
Pittsburgh, PA 15222
Oct. 16-18, 1973: Canadian Computer Show and Conference, East
Annex, Coliseum, Exhibition Park, Toronto, Canada / contact:
Industrial Trade Shows of Canada, 481 University Ave., Toronto,
Ontario M5W 1 A7, Canada
Oct. 16-18, 1973: Input/Output Systems Seminar 73, O'Hare International Tower, O'Hare Airport, Chicago, III. I contact:
DPSA, 1116 Summer St., Stamford, CT 06904
Oct. 18-19, 1973: Computer Science and Statistics: 7th Annual
Symposium on the Interface, Memorial Union, Iowa State Univ.,
Ames, Iowa / contact: William J. Kennedy, Statistical Lab.,
Iowa State Univ., Ames, IA 50010
Oct. 21-25, 1973: 36th Annual Meeting, American Society for
Information Science, Los Angeles Hilton Hotel, Los Angeles,
Calif. / contact: H. W. Jones, Northrop Corp., Aircraft Div.,
Hawthorne, CA 90250

Sept. 10-12, 1973: 5th Congress on Instrumentation in Aerospace
Simulation Facilities, California Institute of Tech nology, Pasadena, Calif. / contact: H. F. S~ift, Materials Physics Research,
University of Dayton Research Institute, Dayton, OH 45469

Nov. 5-7, 1973: Systems, Man & Cybernetics Conf., Sheraton Boston Hotel, Boston, Mass. / contact: S. A. Meer, Signatron Inc.,
27 Hartwell Ave., Lexington, MA 02173

Sept. 17-19, 1973:.. 7th Annual Intergovernmental Council for ADP
Conference, Ottawa, Canada l.contact: ICA Secretariat, 18 Keren Hayessod St~,Jerusalem, Israel'

Nov. 8-10, 1973: 3rd National Conference of the Society for Computer Medicine, Denver, Colo. / contact:
Dr. Joseph M.
. Edelman, Society for Computer Medicine, 200 Professional Ctr.,
244 Peachtree Blvd., Baton Rouge, LA 70806

Sept. 2-S-27, 1973:' Conference on 'Hybrid Microelectronics,' Universi'ty of Kent at Canterbury, England / contact: Registrar,
InstitLItion of Electronic and Radio Engineers, 8-9 Bedford Sq.,
Londop, WC1 B 3RG, England

Nov. 13-15, 1973: Data Networks, Analysis and Design, Tampa,
Fla. / contact: Raymond Pick holtz, Sch. of Engrg., George
Washington University, Washington, DC 20006

46

COMPUTERS and AUTOMATION for July. 1973

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 digital computers
manufactured and installed, or to be manufactured and on order. These
figures are mailed to individual computer manufacturers quarterly 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. A few manufacturers refuse to give out,
confirm, or comment on any figures.
Part 1 of the Monthly Computer Census contains reports for United
States manufacturers, A to H, and is published in January, April, July,
and October. Part 2 contains reports for United States manufacturers,
I to Z, and is published in February, May, August, and November.
Part 3 contains reports for manufacturers outside of the United States
and is published in March, June, September, and December.
Our census seeks to include all digital computers manufactured anywhere. We invite all manufacturers located anywhere to submit inforthat 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
AUTOMATION
C
figure is combined in a total
(D)
acknowledgment is given to DP Focus, Marlboro, Mass., for
their help in estimating many of these figures
E
figure estimated by COMPUTERS AND AUTOMATION
(N)
manufacturer refuses to give any figures on number of installations or of orders, and refuses to comment in any
way on those numbers stated here
(R)
figures derived all or in part from information released
indirectly by the manufacturer, or from reports by other
sources likely to be informed
(S)
sale only, and sale (not rental) price is stated
X
no longer in production
information not obtained at press time and/or not released
by manufacturer

SUMMARY AS OF JUNE 15, 1973
NAME OF
MANUFACTURER
Part 1. United St~tes Manufacturers
Adage, Inc.
'Brighton, Mass.
(A) (May 1973}
Autonetics
Anaheim, Calif.
(R) (Jan. 1969)
Bailey Meter Co.
Wickliffe, Ohio
(A) (Aug. 1972)

Bunker-Ramo Corp.
Westlake Village, Calif.
(A) (June 1973)

Burroughs
Detroit, Mich.
(N) (R) (June 1973)

Computer Automation, Inc.
Newport, Calif.
(A) (April 1971)
Consu1tronics, Inc.
Dallas, Texas
(A) (April 1973)
Control Data Corp.
Minneapolis, Minn.
(R) (April 1973)

NAME OF
COMPUTER
AGT 10 Series
AGT 100 Series

DATE OF
FIRST
INSTALLATION
4/68
1/72

3
7

35
19

1

30
6

0
0

30
6

X
X

8
37
7
15
0
16
1
160
79
15
18
19
19

0
15
0
12
0
0
0

8
52
7
27
0
16
0

0
0
0
2
3
0

1141

677

25

19
23

2
2

1818
500
21
25
600

4.0

2
277

123

400

30

12-14

572

285

857

110

4
154
27

47
8

201
35
60
11

22
2
60

11/58
6/61

Metro type
Bailey 750
Bailey 755
Bailey 756
Bailey 855/15
Bailey 855/25
Bailey 855/50
BR-130
BR-133
BR-230
BR-300
BR-330
BR-340
BR-1018
BR-1018C
B100/500
B200
B205
B220
B300
B1700
B2500
B2700
B3500
B3700
B4700
B5500
B5700
B6500
B6700
B7500
B7700
B8500
108/208/808
116/216/816

10/57
6/60
11/61
2/65
12/72
4/68
3/72
10/61
5/64
8/63
3/59
12/60
12/63
6/71
9/72
7/65
11/61
1/54
10/58
7/65
8/72
2/67
8/72
5/67
11/72
10/71
3/63
12/70
2/68
8/72
4/69
2/72
8/67
6/68
3/69

40-200
40-250
200-600
60-400
50-400
100-1000
100-1000

5/69

0.7

G15
G20
LGP-21
LGP-30
M1000
RPC4000
636/136/046 Series
160/8090 Series
921/924-A
1604/ A/B
1700/SC
3100/3150/3170
3200
3300
3400
3500
3600
3800
6200/6400/6500
6600

COMPUTERS and AUTOMATION for July. 1973

NUMBER OF
UNFILLED
ORDERS

32
12

X

100.:..300

RECOMP II
RECOMP III

DCT-132

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

AVERAGE OR RANGE
OF MONTHLY RENTAL
$(000)
(S)

X
X

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

X
X
X
X
X
X

23.0

X
X

7.0

7/55
4/61
12/62
9/56

X
X
X
X

1/61

X

5/60
8/61
1/60
5/66
5/64
5/64
9/65
11/64
8/68
6/63
2/66
8/64
8/64

X
X
X

X
X
X
X
X
X
X

(S)

2.8-10.0
5.0

23.5
32.0
33.0
30.0
44.0
85.0
200.0
5.0
8.0

X

4

X
X

13

(S)
(S)

2

4
110
225

1
165
215

10
20

175
235

75

65

135
295
20
165
322

X
X
X
X

1

3.8
10-16
13.0
20-38
18.0
25.0
52.0
53.0
58.0
115.0

75
29
610
29
59
428-478
93-120
55-60
205
17
15
40
20
117
88

X
X
X
X

0
C
C
C
C
C
C
C
C
C

47

~~~~i~

~~ IU,RiE~
thWA,
Controt' ilata~cont: d)
",'!,",

'1r)'n

,

cunt d)

N~rA ' i\, 0
L~jAI,"Jlu~

'

'9-~i--

~~'"
yber

- t~iU- -

70/72
Cyber 70/73
Nova
Sup' ernova

Data General Corp.
Southboro, Mass.

N6~~ .0 I
Nova 820
Nova 1210/1220
6024/1
6024/3

Datacraft Corp.
Ft. Laud~rda1e, Fla.

B·•

1MW;mt-~fiNgtMt1\TI13N~

~t~1 ~

.*.

B·• .1\.}.,

,J:.

Wlldd

8
3

5
12
10
4

:.

iii
~7,QQ

(S)
(s)

~~~

III

::

~~~,
H~

1ffiWrnrTI'f
YNFlLLIW
ORDERS
C
C

- -.- .

....'I!iii;

-=-

p,~15

~~~ ~5R

~*~ ~MHt HH~

Datapoint 2200

Datapoint Corp.
San Antonio, Texas
(A) (June 1973~
Digiac Corp.
Plainview, N.Y.
,~~
'lg 7,3)

-IFeb.

2000

-- 'llU
L"" a~~8

l'f,7,O::

B::IW'

~(j---

,,1'-

B1~tHL -Effili~mH--f8Rff81~ I -HW:

1/7~

D-1l6

airfield, N.J.
(A) (June 19732
Digital Equipment Corp.
Maynard, Mas s.
(A) (May 1972)

11/60

1'D1'-l
PDP-4
PDP-5
PDP-6
PDP-7
PDP-8
PDP-8/1
PDP-8/S
PDP-8/r,

8/62
9/63
10/64
11/64
4/65
3/68

11/66

p~~;~Z!

11/68

fll?B=~/F

5/72
12/66
11/68
12/67

f,H?~~/M
1l~1l.,,1J

!3B~::~t
~~~§y§ tem-1C!
fbP-1JlfO

llDP=;LlR2Q
rpl'~ll/OS

PDP-ll/45
PDP-12
PDP-15
LINC-8
640
8400
PACER 100
SPC-12
SPC-16
S;tstem 18/30
GE-PAC 3010
GE-PAE 4q1q

~leetronic Associates Inc.

West Long Branch, N.J.

~A~ (June 1973)
General Automation, Inc.
Anaheim. Calif.
~A2 pune 1973)
General Electric
West Lynn, Mass.
(Process Control Computers)
(A) (June' 1973)'"
),T',nJrers;

,-

9/69
2/61
9/66
4/67
7/67
7/72
1/68
5/70

3169
" '"1nW

~¥.~r~~ 4M~

~'I:!¥!-o, ~~~g

-

-

,

--

~13~¥~~ 40,5()
GE-PAC 4060
"'Z114B .

~n5A

2116B.

10.0

(iT)

1.5.5

(9)

793

140
84

895
877

48
40
90

2
5
10

50
45
100

,-

=

lQO
l4Q2
3H7

lQIO
X

X
X
X

700-3000
10.8
13.8
10.8
17.0

X

9.6

X
X
X
X

~

6/70

8/68
1/66

¥:=,l,25
ij..,iOQ

12/67
3/64
12/61
12/60
2/66

X
X

~~

i\
X

!HO

X
X
X
X

3699
3787
365
2
436

X

40

X

X

~OO

0
0
620
545
200
170
.29
75
1900
1250
375
26
~4

X

1
0
25

X

:32.

f60

X

65
25
20
1182
333
1171
4
6

X
X
X

200-400

420-680

620-1080

11
35
15
145
40-60
3
15-20
~3=Bg

0
0
1
15
17

11
35
16
160
57;"77

X
X
X
X
X
X

?C

1?=99

§P-90
10
15-45

70-100

240-400

310-500

50';'100

20-30

70-130

::
lQ.,,~Q

X
X

],,0

25.0

~...,12.0

U"'l~OO

X

14.0

R..U~
U-=~OO

X
X

7/69
3/68
4/65
5/65

R..,800

1.0
1.3
1.4
1.2
2.2
2.9
4.5
X

Q!~~
~/~5

(;625
~~

\!,.".UO

1
4
69
20
2
2

0.41
0.6

7, :J
23.0

O&B

25
3Q

0.25

6/6.7

GA4Q WS

61
8
35

:IN!?6

6/64

G4~0 T/~
G,4~~

109
21
40

(S)

45

9:§

~42g T/:~

0
0

23
18

5/6.4

%?s,

0
0

X

~

8
1

14

(s)

(S)

It!

20

243
';.740

(S)
(S)

::

=

-X - -

T(~

§iB

(9)

X

1.2
12.0
1.0

yg~
nfg~

(;lQ~~
mQ~R,\,~
GU~

(II)

X

T/S
T/S
T/S

moM;\.

(~)

j{

U!1'I7
11 66

0

a

=

X
X
X
4.9
3.9
3.9
X

78
~g

X
X

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

~5,~'-

GUO
elSO
G205
G210
G215
G225
G235
G245
G255
G265
G275
G405
G4l0

(S)

g':"f1

6/1'15
lQ/Q8

all

9.0
9.0

ttH1Q
Uf37
~p6~

-?n4A~

21M.

M\

'

~n

-1~J
9.2
9.6
~;~'
6.9

NOV~1~~~8

f~~ Hftf: 'l~H~

III- -,
,:, 'i"~~~Mf ",' t

"-WM#f~~
B4!f~ff. §

U.O
20

6

23
20-40
180
30
800
150

3
3
7

26

32.0
X

47.0
2.7
3.5
4.8
7.0
7.5
10.5
30.0
9.8

BOO
46
57
230

160
220
275
40
15
90

26
23-43
187
960
370
1075'
86
72
320

X

0

X
X

COMPUTERS and AUTOMATION for July. 1973

ilia

:I

.n

Sf

N~
~H~f\fJmHH~

B~t~ OF

N~·pF

F8}.l¥BH~
H-1250
H-1400

tr9~~w~n~r=9f1rri!~

H::H99
~~t9t~
Jl-20 P

n~~T

Hl§IM:HH8~
7/68
1/64
1/64

=

":'

H::2050

ftH:~~~
t,
J!:::;~P8

!keggg
tkti~~§

BB?=?

1

6
5

=
=

10
20

~

x

':

:-

--

60

~
~

ti~

---

~3

13

=

2~~

*

60
452
350

x

l:l:§:ft:

l

l!i

::-

=

4/6

3/§3
6/69

=

0-.6

~/ee

l;±

=
=

=
:=

=

=-

~
~

x
x

10/69
12/68

3.2

11/68

12.0

i~ J; th~

PROBLEM 737: CUBe DIVISION

When the boss entered. the Computer. Center, he· saw
§@y~!~l people drawing geometrical figures on scrap pads.
"What's all this?" he asked.
"It's a probiem'Pete is writing a program for," said Joe.
"We're trying to beat him to it with R~p~n fW§ p~per."
~'WpM FF8BI~rn is th~t?"
.
.
nT..p.. PH!. i! §imRJ~; if;~ fa .nmJfhc mff*imHm flHffiBrF 8f
HJt>~& ~ eYRe EfHft HE g~vided into.'?
ffe~n?t f?e 4jvig@9 !nW?Wh~! ~ina gf 40uple talk is
th~t1B
m~

pegin with squares since that?s hmv F@t@&8!

§tAftgg gn thj~~ H§ found that a square could not be divhj~Q into fiv@ §maH§f§ijHflf@§j 9Htf9H!9 g@ eHt u~ ~!1!9
six, RQVen or iHlY· W@at@fnHmgef gf §~Hafe§:n
HUQW gan a §Q~are b~ cut· up into six Qf §~¥en§maHef

squurcs?B
"Well, for six you can cut a square into nine smaller
squares and imagine the four in one corner to coalesce
into one larger square." Joe drew a figure on his scrap
pad. "For seven you could cut a square' into. quarters,
th~.~ cut one of these pieces into. quarters." .He drew
~r.q~P~f H~Nf~ tR !~lustrate this.
."',~ng g~!,~ i~ UYip& fR 88 fh~ same thing for cubes,

.

ynu

-.

~r???

~~ygs? fi~?§

"-'

twt a Jgt &f ifl§tnlSHgfl§f~H EHHiH~ HF EHB&~?

gmlh~§gin~ them

Rml d9jn~otner 8?efffH8fl§:H{thfr.~fh~'
Hli§ tggeHlef ami SOllt HpwHh fht ~nsw~r.
But I'm pfatty §Yf§ Weill §S8ml him en thl§:~~
What is. the maximum num~~f Rf EHBe~ inf8 whiEh .~

linn

~ytaH

cube can not be divided?

§9!Ytign f9 ~f8e'~m l~R;§w@~f1in~ NHmOOF~

the fif§! flHllB@f i§ ~, !h~
nine H~maiIliH~ ~JfJ~e§? with !he

If HH~f@ anr tWg §We~IJ§ gHd
1 can b@ in {my of th§

COMPUTERS and AUTOMATION f!~fJyIYI

8!ffiH~

-18$- _.

X

-:

,.,

X

~50

X

x

~gg

73

=

.,.

20

nymR@f§ .~ to 10 appearing in order. If the first number

WnltorPenlley, QOP
Problom Editar
Computcmi and Ayt@m~Hhm

Hk§t

_H

OOfftHB

12

Ht§~~~

PROBLEM C:ORN6R

f.'J8ftH

~

10

~;M

DPf=~g~

DDf=316
m)p.,.416
mW::§l§
H112
H632
IH602
H1642
H1644
H1646
H1648

NUMBER OF

4
15

50.0

3/6~

In

1$0

I3I3f=He

DDP-124

811 W§Htt~H8N~ .

12.0
14.0
50.0

12/68

3pe~

In

8HE~tH§
H:§:A:
---55'- ..

~t~

=

- .. NU~f:tt

8f M8 Hl:¥ ftfNHt
§ 998~

il~i=

H-60 0

!k:§~g~

~VERAGE~gJt -~p~

19'7§

Of(~)= 36 places, the
1 gggYffin~ in gni@f: Thm th~r~ af§ m9f~ arran~8m~nt§
producing two sweeps with inHifl! 4 than with ~ftHia.l ?:
Similarly the number increases to initial .s, remaining can.,.
l'!I1g

7 may occur in

any

stant for initial 6, then decreasing.
Across the

Editor~s Desk

- Continued from page 43

The .work will be performed by Informatics Inc.
terns and Services Company, Rockville, Md.

sys~

Under phase one of the contract, Informatics will
an~nternational Cancer Re.,..
~~~Fch h~fHg~~~, ~htfh ~tJJ fgtJ9ct, st~~e a~d dISsemInate results orcanc~r ~eSea~ch toS~~lentlsts
throughout the world; (2) cancer Ii teratureand data
services such as ind~xing, abstracting and translat.,.
ie g ; and (3) information support services, which

R~~tn ~eve~pp~ ng:

O!

wil1proqHP~tnfwm9H8n ~H8n ~~ air~f~8f}~g, gNf:

veys ,progress reports ,catalogs ana speCIal teports.

§P~El~lhRF8J9Bf~ ~8.b~1~"~~F~e~~Rll~ }e~ fl.iif~f

P. h
.• e~~ Sf t _~ c'6~tr~p} In~ ~9.~ a cata og OL a
InformatIon systems and ser~H~escurJ'entlybeingof­
feredby NCI, .and ,a directory of organizations and
ingividu~l sthroughout the world offering cancer research and treatment. The CIS isa~re;]dy working wi th
the Institut Gustav~Roussy in Pari§, which ~~gh
month reviews cancer literature from 200Q Jgurn~l§
and ,distributes the information in three 1an9uog08
to five countries. Cooperation with other organizations, such as the Association of American Cancer
Institutes, the National Library of Medicine, the
Regional Medical Libraries, and services such as
Biological Abstracts,Chemical Abstracts, Current
fSlli~rrfJ ~n8 G'f~IJ1l~. Me~~ca, will be explored.
I

__

Under the National Cancer Act of 1971, specific
responsibilities were assigned to t.he Dir,ectorof
~Ef for the collection, analysis, storage and diss,mi~~tibH~o~UA~hcet research information, nations r.plnat~OlJ f C;i(,':f
Th I
.
1 Cancer
il.t
Y llfln1n er:natlonally. . e nternatlona
·n~g~~#RB~ X~~r~ )n~~ peen authorized by the Con~
ijf~llt8 mB~~ ~~~tlable to any cancer researcher 1n
§ny.country the results of cancer research efforts.
A unified and coordinated undertak~ng of this kind
I~ new not only to cancer research, but to the field
o£biosciences itself.

t

49

COMPUTER-ASSISTED ANALYSIS AND
DOCUMENTATION OF COMPUTER PROGRAMS
•

•

ARE YOU INTERESTED IN
THIS KIND OF PROBLEM?

Problem: You have a piece of software, a working
binary program (WBP) In machine language -- with operating instructions, and it is useful in your installation -- but you have no idea how it works in
detail and you have no idea how to modify it to suit
new conditions or requirements. (The programmers who
wrote it went away two years ago to their next job -or they never worked for you at all -- or ... )
But YOU have the problem of understanding and modifying that program, salvaging what you can of it;
YOU have the problem of making its subroutines useful in other programs by calling them when you want
them; etc.
Theory: Use a computer to assist you in your detective work analyzing that program (the WBP) and producing documentation for it. Desired Goal: Complete
understanding.

IF THIS KIND OF PROBLEM IS INTERESTING TO YOU,
WHY NOT TAKE A LOOK AT THE FOLLOWING BOOKS?

Research in Computer-Assisted Documentation
of Computer Programs

by Edmund C. Berkeley
principal investigator, author
- Vol. I, published by Information International,
Boston, Mas s. ,
softbound, April, 1969, 128 pp, $3.00
- Vol. 2, published by Berkeley Enterprises Inc.,
Newtonville, Mass. 02160,
softbound, Nov. 1971, 112 pp, $3.00
Technical (but understandable) reports produced
and research done -- under contracts with the Office
of Naval Research (N00014-68-C-0268, N00014-C-70-C0225)
Volume 1, CONTENTS:

Idea 1: Use a computer program (SIMULATOR ANALYZER)
which will simulate your computer, and operate the
given program (WBP) on examples, step by step, from
one point to another point, showing you intermediate
results, telling you where control goes.
Idea 2: Use a computer program (RELOCATOR) which
will shift the WBP into another area of core, and
thereby reveal which machine words can be moved unchanged, and which machine words have to have the
shift difference added or subtracted from them.
Idea 3: Use a computer program (SUBROUTINE EXAMINER)
which will show how each subroutine in the WBP operates on each kind of information that comes into it.
Idea 4: Apply techniques of CRYPTANALYSIS to discovering what systems of character representation are
being used in the computer program.

- The Subject and Purpose of this Research
The Documentation of Computer Programs
Some Estimates of Loss Due to Inaccessibility
of Computer Programs
- The Concept of the Simulator Analyzer
- Model 1 and Model 10 of the Simulator Analyzer
- Cryptanalysis of a Portion of a Computer Program with Unknown Documentation
Volume 2, CONTENTS:

- Simulator Analyzer Model 13
- "Comments" in Computer Programs: Principles
for Abbreviating, and Suggested Abbreviations
- Successful Relocation of the Working Binary
Program for "Old 16K DDT" (Dynamic Debugging
Program) Without Knowing the Symbolic Program
from Which it was Assembled

Etc., Etc., Etc.
RETURNABLE IN 7 DAYS FOR FULL REFUND IF NOT SATISFACTORY (IF IN SALABLE CONDITION)
WHY NOT TAKE A LOOK? . _.. HOW CAN YOU LOSE?

(may be copied on any piece of paper) - - - - - - - - - - - - - - - - - To: Berkeley Enterprises Inc., 815 Washington St., Newtonville, MA 02160
Please send me both volumes of "Research in Computer-Assisted Documentation", for $6.00.
I already have Volume 1; please send me Volume 2 only, for $3.00.
I enclose $___ in full payment.

) Please bill me.

) Please bill my organization.

RETURNABLE IN 7 DAYS FOR FULL REFUND IF NOT SATISFACTORY (IF IN SALABLE CONDITION).
Name_______________________________________________________________________________________________

2
2

Address _____________________________________________________________________________________________
City______________________~________________ State ___________________________ Zip_____________________

2
3

Signature_____________________________________________ Purchase Order No. _____________________________

50

COMPUTERS and AUTOMATION for July. 1973

c

ttWHO'S WHO IN COMPUTERS
AND DATA PROCESSING"
EDITION 5.2

= 5th

EDITION + 2 SUPPLEMENTS

THE MOST ESSENTIAL COMPONENTS IN COMPUTERS AND DATA PROCESSING ARE PEOPLE

Who are they?
What do they do ?
Where do they do it ?

ANDREE, Richard V. / professor, author, lecturer,
consultant / b: 1919 / ed: BS, U niv of Chicago,
PhD, Univ of Wisconsin / ent: 1948 / m-i: A Ma
P Sy; writing, information science -/ t: professor of
math, research assoCiate in computing science /
org: Univ of Oklahoma, Norman, OK 73069 / pbh: ACM, AEDS, ASL, DPMA, MAA, NCTM,
SIAM lecturer, American Assn for the Advancement of Science, American Math Society, American Society for Engineering Education, Mu Alpha
Theta, Pi Mu Epsilon, Sigma Xi, 3 fellowships,
numerous committees, Who's Who in America,
World Who's Who, editor, 12 books, 8 paperbacks,
about~20 articles / h: 627 E Boyd, Norman, OK
73069

Consult
"WHO'S WHO IN COMPUTERS
AND DATA PROCESSING"
jointly published by Quadrangle Books (a New York Times
company) and Computers and Automation (Berkeley
Enterprises, Inc.)

We are confident that you will find the'subscription
will repay you many times over. In fact, one day when
this wealth of material gives you the inside track with
someone important to you, you'll find the information
PRICELESS: the most essential component in EDP is
GOOD PEOPLE.

"Who's Who in Computers and Data Processing" has
been changed to an annual subscription basis as follows:

1. The latest Cumulative Edition (the 5th edition
published 1971, containing over 15,000 capsule biographies, over 1,000 pages long, 3
volumes, hardbound) PLUS

RETURNAaLE IN 10 DAYS
FOR FULL REFUND
(if not satisfactory)

2. At least 3 Updating Supplements per year expected to total over 3,000 entries
Both for $34.50 per year in any year when a cumulative edition is supplied ... and $15.00 per year in any
year when a.cumulative edition is not supplied.
BASED ON "Computers and Automation" 's continual data
gathering from computer professionals

r- - - - - - I

(may be copied on any piece of paper)- - - - - - - - - I

WHO'S WHO IN COMPUTERS AND DATA PROCESSING
815 Washington St., Newtonville, Mass. 02160
YES, please enroll me as a subscriber to WHO'S WHO
IN COMPUTERS AND DATAPROCESSING at
the following rate:

I

$34.50 including the last cumulative edition

This reference is particularly useful for:
Personnel managers
Libraries
Conference planners
Directors of computer installations
Suppliers to the computer industry
Execll t ive search organizations
Prospective authors
Prospective speakers .. ,

OR
$15.00 since I already have access to the last cumulative edition
I understand that in each .12 months I shall receive at
least three updating supplements, expe~ted to total over
3000 entries.
Payment enclosed
f Please bill me.
) Please bill my organization
RETURNABLE IN 10 DAYS
FOR FULL REFUND
(if not satisfactory)

anyone who needs to keep up with the important people
in the field.
.
Each computer professional has a capsule biography
detailing: last name; first name 

CD

CD

:l

J:

...

..
e

~
.r.

+"

"61

0
Q)

Q.
co
1;;

I

m

..J

~

~

>
..J

c..
w

m

>-w

<2:

-.J
W

W

a:

co

co

E

>-co

:=

0

~

;;{
a..

:::CD

w

u

CD

C

Co
+"

Cii

0

CD

co

co

III
Q)

Cl

co

+"

III

0
Co

0

Z

Z

Z
w

"'C

en
en
:::>

a:
a..
a:

:I.!

E

Z

(J)

c

a:

w

w

w
I-

co
-.J
-.J

~
w
CJ


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                     : 2009:03:25 00:58:06-07:00
Create Date                     : 2009:03:25 00:58:06-07:00
Metadata Date                   : 2009:03:25 00:58:06-07:00
Format                          : application/pdf
Document ID                     : uuid:3fd95695-3551-420e-88ac-d7c125c9a0c3
Instance ID                     : uuid:53882afa-6ab5-4768-90b1-fa9163062ef6
Page Layout                     : SinglePage
Page Mode                       : UseNone
Page Count                      : 52
EXIF Metadata provided by EXIF.tools

Navigation menu