196604

196604 196604

User Manual: 196604

Open the PDF directly: View PDF .
Page Count: 64

Download
Open PDF In Browser	View PDF

AN JOSE
PUBLIC LI BRA RY

_:="lA'
•
Glass master negative for producing printed circuit boards for computer",

April, 1966

Clark Equipment Company gets data
from 127 sales offices, 4 manufacturing plants,
and a major warehouse as soon as it's recorded

Bell System communications is the vital link

Bell System data communications services link
Clark's distant locations to a centralized computer
center at Buchanan, Michigan. The result is better
management control of all activities-sales, inventory,
purchasing, production, payroll and accounting.
With current and accurate information, Clark management can quickly adjust to changing marketing
conditions. Important orders get priority scheduling
for production and shipment. And yet, purchasing,
production and inventories stay at optimum levels.
An integrated information system of this size uses
computer switching with store and forward capabilities.
The fully automatic Clark system polls satellite stations,
receives and transmits messages, assigns priorities,
and converts different speed and code formats to

See the latest in data communications.
Visit the Bell System exhibit at the Spring Joint Computer Conference.

one standard code. Other features of the switching
unit provide the necessary supervisory control of the
network.
Consider the economies a real-time, integrated
information system can bring to your business with
automatic data processing linked with fast, reliable
communications.
Today's dynamic competition requires many companies to consider organizing for data processing in
some phase of their operations. It's important to start
organizing communications at the same time.
So when you think of data communications, think
of the Bell System. Our Communications Consultant
is ready and able to help you plan an integrated
information system.

@ Bell System
N.t1I.
•

American Telephone & Telegraph
and Associated Companies

I
!
I

I
it

IN SCIENTIFIC
PROGRAMMING?
At 0 he d, opportuniti s in the fi d of
scientific mputer rogramming a ound. nd
e f assignments is
int esting
it
is b oa· : in
hysic sciences, engin
dy amics, 0 mand
. contr simul · n
advanced s ,
programs. hat
0
s
of the
is why, to this
world's largest industrial
uter
ckheed
committed
ti
c
s y s , nd clear, t 0,
th
at d opportunities that a wait exp
scientific progr
s at Lockheed.
:Please write to lVI.r@ ....
id.doo, Pro.fessional
lacement
ag
u
ale, Californi .
L
he is a equal opportunity e
oy,
$

LOCKHEED
Designate No. 4 on Readers Service Card

COMPUTERS and AUTOMATION for April, 1966

MISSILES & SPACE COMPANY

Mostpeople
IikeComputape

A few don't
The way repeat sales are going lately, there are an awful lot of people out
'
,
there who like Computape.
But occasionally we run into someone who doesn't.
Bound to happen, of course. Once in a long while it turns out to be someone who has a legitimate gripe. Like the little car ad says, nobody's perfect.
Muchmore often, interestingly enough, it's someone who has never even
tried Computape. Maybe he's found another brand that seems adequate and
would rather fight than switch. Or maybe he has a. feeling that the company
that does the most and the loudest advertising just naturally makes the best
precision tape.
We will respect his opinion without subscribing to its validity.
Nevertheless, we would like the chance to prove to him that Computape
is the finest, most dependable tape that money can buy. Tape is our only
business, so it jolly well better be.
Maybe you're missing out on something good, too, just because you've never
tried it. Why not investigate? After all, most people like Computape.
A PRODUCT OF CoMPUTRoN INC.

"Visit our exhibit ot AFIPS Booth No. 1202-1203."

MEMBER OF THE

Designate No. 5 on Readers Service Card

rntMllr GROUP

122 CALVARY ST" WALTHAM, MASS, 02154

The frollt cover shotlls
a glass master negative !tsed to produce
cirettit boards for the IBM System/360
For more in/ormation see page 51.

-1

.Jr.

computers and data processors:
the design, applications,
and implications 0/
information processing systems.

editor and publisher
EDMUND C. BERKELEY

auociate publisher
PATRICK J. MCGOVERN

I),

'"r")'

., !

auistant editors
MOSES M. BERLIN

LINDA LADD LOVETT

contributing editors

"'I

{) j

.--1 ( ) , , ....

P. Snow

THE AUTOMATED PROCESSING OF PEOPLE
The problems of automated physical transport of a mass population
and the merging of transport, communications and computers, to obtain
more "usable" lifetime.

DICK H. BRANDON

THE HUMANITIES AND THE COMPUTER: SOME CURRENT RESEARCH PROBLEMS
by Edmund A. Bowles

ALSTON S. HOUSEHOLDER

The advantages of the computer to the humanities: liberation from chores
. and seven league boots for analysis.

PETER KUGEL
ROD E. PACKER

by Rod E. Packer

JOHN BENNETT

NED CHAPIN

) :' J! : ;:,'

The relations of science and revolutions to rich and poor societies, the
role of the computer, and what people can and should do about it.

ANDREW D. BOOTH
JOHN W. CARR, III

••

SCIENCE AND THE ADVANCED SOCIETY
by

NEIL D. MACDONALD

USING A COMPUTER TO DESIGN A COMPUTER
by Kathe Jacoby and Armand R. Laliberte

An explanation of and an argument for "design automation" of computers in the design and production of new computers.

adviJOry committee
T. E. CHEATHAM, JR.

the use

JAMES J. CRYAN
RICHARD W. HAMMING
ALSTON S. HOUSEHOLDER

1~'CI'OSS

HERBERT F. MITCHELL, JR.
VICTOR PASCHKIS

art director

WILLIAM J. MCMILLAN, 815 Washington St.
Newtonville, Mass. 02160, 617·DEcatur 2·5453

Perspective

l.'1:1r/2ct rejJOrt

RAY W. HASS

fulfilment manager

Burroughs Enters Two in the Third

tbrougb/wl
30

by Dick H. Brandon

C':I)!t.:!
advertising, representatives
New York 10018, BERNARD LANE
37 West 39 St., 212·BRyant 9·7281

Chicago 60611, COLE, MASON AND DEMING
737 N. Michigan Ave., 312·SU 7·6558
Los Angeles 90005, WENTWORTH F. GREEN
300 S. Kenmore Ave., 213·DUnkirk 7·8135

9
59

Elsewhere, THE PUBLISHER
815 Washington St., 617·DEcatur 2·5453
Newtonville, Mass. 02160

rCj)ort

by James Titus

redders' Iwd editor's /orulll
9
9
9

San Francisco 94105, A. S. BABCOCK
605 Market St., 415·YUkon 2·3951

,'be cdi/oi"S des/!.

COMPUTING AND DATA PROCESSING NEWSLETTER

International Computation Centre - Comments
Annual Computer Art Contest
The June Issue of "Computers and Automation": Computer Directory and Buyers'
Guide
Correction
Calendar of Coming Events'
*D ON YOUR ADDRESS IMPRINT
MEANS THAT YOUR SUBSCRIPrej..:rc))cc hl/ormatiolJ
TION INCLUDES THE COMPUTER
Computer Census
DIRECTORY.
*NMEANS THAT'
New Patents
YOUR PRESENT SUBSCRIPTION
DOES NOT INCLUDE THE COMindc.,· of notices
PUTER DIRECTORY.
Advertising Index

:~~

COMPUTERS AND AUTOMATION IS PUBLISHED MONTHLY AT 815 WASHINGTON ST., NEWTONVillE, MASS. 02160, BY BERKELEY ENTERPRISES, INC. PRINTED IN U.S.A. SUBSCRIPTION RATES: UNITED
$15.00 FOR 1 YEAR, $29.00 FOR 2 YEARS, INCLUDING THE JUNE DIRECTORY ISSUE; CANADA, ADD SOc A YEAR FOR POSTAGE; FOREIGN, ADD $3.50 A YEAR FOR POSTAGE. ADDRESS ALL
••••••••• EDITORIAL AND SUBSCRIPTION MAIL TO BERKELEY ENTERPRISES, INC., 815 WASHINGTON ST., NEWTONVILLE, MASS., 02160. SECOND CLASS POSTAGE PAID AT BOSTON, MASS.

.Y!J[.' STATES,

POSTMASTER: PLEASE SEND ALL FORMS 3579 TO 8ERKELEY ENTERPRISES, INC., 815 WASHINGTON ST., NEWTONVILLE, MASS. 02160. © COPYRIGHT, 1966, 8Y BERKElEY ENTERPRISES, INC. CHANGE
OF ADDRESS: IF YOUR ADDRESS CHANGES, PLEASE SEND US BOTH YOUR NEW ADDRESS AND YOUR OLD ADDRESS (AS IT APPEARS ON THE MAGAZINE ADDRESS IMPRINT), AND ALLOW THREE WEEKS
FOR THE CHANGE TO BE MADE.

COMPUTERS AND AUTOMATION, FOR APRIL, 1966

Automated
Payroll
for
LESS Money!
Now, you can electronically automate
complex payroll applications (plus other
processing jobs) for substantially less
cost than any other type of automated
EDP system.

"500" payroll applications can be
TOTAllY automatic. It provides each
employee with a fully protected check
and an itemized statement of earnings
and deductions. The payroll journal,
check register and employee's earnings
record are all produced automatically.
So is a distribution detail for direct entry
to subsidiary ledgers. Even details for de·
ferred records like quarterly tax reports
are accumulated and stored for subse·
quent automatic reporting.
The NCR 500 System reads and "speaks"
in any of four EDP languages. Punched
card. Punched tape. Magnetic ledger
Optical character recognition. You can
use any one-a combination, or all of the
languages with equal ease.
Buy or lease only the units of the system
you need. Expand it as your business
grows. The price of the "500" System
will remain modest.
What can the NCR 500 System do for
you? What are the probable savings? You
can find out by calling your local NCR
office. Or, write NCR, Dayton, Ohio 45409.

THE NATIONAL CASH REGISTER COMPANY

Designate No. 6 on Readers Service Card

DAYTON, OHIO

45409

COMPUTERS and AUTOMATION for April, 1966

c&a
EDITORIAL

Perspective

In this issue we publish an off-the-trail article by Lord
C. P. Snow. His article "Science and the Advanced Society"
helps us see our world in better perspective. He talks of a
number of important, sharp changes, revolutions, happeninR
in our world. The reason we publish this article - although
it talks about computers in only a few paragraphs - is that
in general in our society:
Too much time, energy, and thought are devoted to. less
important problems and not enough to more important
problems.
What is the evidence for this proposition? How do we
measure the importance of problems? And if this proposition is true, what should we do about it?
In every society some problems are important, some problems are trivial. The importance of a problem in a society
is measured by a number of factors. Among these are:
• The number of people whom the solution will affect
and the degree to which they will benefit - the value
of the solution;
• To what extent the solution of a problem enables other
problems to be solved - its fruitfulness;
• The nature and quantity of work and resources required to solve the problem - the possibility of solution and the cost of solution;
• To what extent a new and better solution can bring
improvements over prior solutions - the differential
gain; etc.
In many cases nowadays, of course, the possibility of solutions, the cost of solutions, and the degree of improvement of
a new solution over prior solutions, are greatly affected by the
programmed computer.
In the segment of society called the computer field, a
great deal of time, energy, and thought is nowadays going
towards such problems as:
• The design of computers using integrated micro-circuits so that higher speeds and lower costs will be
attained;
• The development of new and better computer programming languages so that more powerful methods
for solving problems can be more easily used;
• The spreading of the use and application of computers into more and more organizations, so that the
advantages of computers can be grasped in more and
more organizations.
COMPUTERS and AUTOMATION for April, 1966

But in society there are much more important problems
- and much harder to take hold of. One of these is the
process for gathering data which the people of a society
have for solving their problems. In other words, this is the
problem of dealing with the stream of information and/or
propaganda provided by newspapers, radio, television, etc.
This problem is highlighted by such an example as the
many flatly contradicting statements issued day by day by
the government and the press in the U2 plane incident of
May, 1960, and the Bay of Pigs incident of April, 1961.
Another of the great problems is that of food or starvation
for tens of millions of human beings in the underdeveloped
countries. This is highlighted by Snow's statement that a
million inhabitants of India are expected to die of' famine
this spring.
Still another of the really important problems is the problem of the lack of control over nuclear weapons. These are
now in the possession of at least five nations: the United
States, the Soviet Union, Great Britain, France, and China.
This problem is highlighted by the fact that the nuclear
stockpiles of the United States and the Soviet Union are
each sufficient to eliminate all human life in the Northern
hemisphere.
Computer people, as we have said before, are essentially
information engineers. They have hold of what Snow calls
the most remarkable machine invented by man while on
this earth. To gain the greatest value from our magical,
pearl of wisdom, we need to look at the engineering of information for human purposes in the broadest possible way.
We need to see both big problems and small problems; we
need to be able to tell the difference between them; and we
need to devote solid work to the big problems - even if
our only reward for a long time is abuse and punishment.
We need perspective.

c.~
EDITOR

Literally translated from the Latin
the headline means a "new breed."
Literally speaking, that's the most
honest way to describe today's
Manufacturing Engineer at IBM
Kingston. It's here we manufacture
a product line ranging from Precision
Display Systems to High-Current
Power Supplies; Miniature Ferrite
Memories to Large-Scale TimeSharing Systems.
To accomplish this successfully the
requirement is for a "new breed" of
Manufacturing Engineers; individuals who can create new techniques,
innovate new procedures and work
in previously unexplored areas.
Individuals who can move across

such diverse disciplines as chemical
milling, linear programming, electron
beam welding, automatic circuit
testing, automated assembly,
computer control of manufacturing
processes and dozens of others.
If you believe your training and
abilities fit you to make important
personal contributions to totally
new manufacturing systems, we
suggest you contact us immediately.
There's lots of room to move your
career forward because we anticipate
introducing twice as many technological changes in the next 5 years
as we have in the last 15.

Immediate openings are currently
available to individuals possessing

degrees and/or equivalent experience
in these areas:
o Systems Planning 0 Programming
o Automation Engineering 0 Process Control 0 Process Engineering
o Plastics Engineering 0 Chemical
Engineering 0 Printed Circuit
Engineering 0 Mechanism Design
o Metallurgy 0 Test Equipment
Engineering 0 Logic Design 0 Circuit
Design 0 Package Design
o Tool Design
Direct your inquiries in complete
confidence to: Mr. C. E. Nelson,
Dept. 539R, IBM Corporation,
Neighborhood Road, Kingston, N. Y.
An Equal Opportunity Employer
(M/F)

Designate No. 7 on Readers Service Card

COMPUTERS and AUTOMATION for April, 1966

c&a
READERS' & EDITOR'S FORUM

INTERNATIONAL COMPUTATION CENTRE
-COMMENTS
C. Berge
The Director
International Computation Centre
Rome, E.U.R., Italy

I would like to thank you for your excellent and helpful
editorial about ICC in "Computers and Automation." This
was of great help to us and we have already received two
letters and small donations from people who read your article.
I. would also like to tell you that your book "The Computer Revolution" has been widely circulated among our
students.

THE JUNE ISSUE OF "COMPUTERS AND
AUTOMATION": COMPUTER DIRECTORY
AND BUYERS' GUIDE
The June issue of "Computers and Automation" is the
annual Computer Directory and Buyers' Guide which we
publish. This issue contains:
• Roster of Organizations in the Computer Field
• Roster of Products and Services: Buyers' Guide to
the Computer Field
• A Special Roster of Electronic Computing and Data
Processing Services
• A Special Roster of Consulting Services in the Electronic Computing and Data Processing Field
• A Special Roster of Software Suppliers

ANNUAL COMPUTER ART CONTEST

• Characteristics of General Purpose Digital Computers
• A Survey of General Purpose Analog Computers

As in previous years, the front cover of the
August issue of "Computers and Automation" will
present the first prize in our annual computer art
contest.
Any interesting and artistic drawing, design,
or sketch made by a computer, analog or digital,
is eligible. It should be provided if possible on
white paper in black ink, so as to make a good reproduction; but this is not indispensable. The way
in which the
drawing was
made by a computer should be
explained. There
are no formal entry requirements:
any kind of letter submitting
the design is
acceptable.

Computer Art First Prize, August 1965

The deadline
for receipt of
entries to be
considered in the
contest is Friday,
July 1, 1966.

COMPUTERS and AUTOMATION for April, 1966

• A Survey of Significant Special Purpose Computers
• A Roster of School, College and University Computer
Centers
• A Roster of Computer Associations
• A Roster of Computer Users Groups
• A List of Over 1000 Applications of Electronic Computing and Data Processing Equipment
Subscribers of "Computers and Automation" are of two
kinds: subscribers receiving the Directory, whose basic subscription rate is $15.00 a year; subscribers without the Directory, whose basic subscription rate is $7.50 a year. To tell
which kind of a subscription our records show for you, look
at your address label on the outside cover; if it has a *D,
our records show that you receive the directory; if it has a
*N, our records show that you do not receive the directory.
A nondirectory subscriber may become a directory subscriber by' requesting the change and sending us $7.50 for
the directory issue. The price of the directory issue to nonsubscribers is $12.00.

CORRECTION
In the February 1966 issue, on page 40, the story entitled
"COMPUTERS BUILT INTO COMMUNITY HEALTH
PROGRAM" was mistitled. It should have been entitled
"COMPUTER SELECTS COLLEGES FOR HIGH
SCHOOL STUDENTS."
Please correct your copy.

MARKET
REPORT
BURROUGHS ENTERS TWO IN THE THIRD...
Late last month Burroughs announced two new microcircuit computers as its latest offering in the third generation computer market. The two new processors, the
B2500 and B3500, incorporate the key features of third
generation computer equipment; i. e. (1) modularity,
(2) compatibility, and. (3) expansibility.
The new processors are described as first members
of a B500 series, and it is expected that additional members of the new processor family may be announced soon.
The design philosophy of the B500 series seems
similar to IBM System/360, but Burroughs President
Ray W. McDonald insists that the new systems are "definitely not carbon copies" of systems being marketed by
principal competitors. According to McDonald, "Our new
syste'ms provide a greater 'throughput' ratio and a higher
degree of responsiveness than any competitive system
series. We have proved this in our laboratories, and we
will be proving it very clearly to customers when deliveries begin in about nine months. "
The new processors are priced to compete in the
small-to-medium-scale computer market, with workable
configurations renting for $4500 to $20, 000 a month.
The 2500 and 3500 offer a memory capacity ranging
from 10, 000 to 150, 000 eight-bit bytes and a wide range
of peripheral devices. The core-memory cycle time for
the 2500 is 2 usecs. for each two bytes accessed. For
the 3500 it is 1 usec. per two bytes. The overall performance of the 2500 is rated by Burroughs as between one
to one and a half times that of the IBM 360/30 for a somewhat lower price. First delivery of the 2500 is scheduled
for January, 1967, and for the 3500 in May, 1967.
Simultaneity All peripheral operations are independent of each other and of the central processor. Any combination of simultaneous input/output and processor operations is possible. Each peripheral device competes
along with the processor for use of the core memory.
Accesses are controlled by a priority system through a
central controller. In the event of multiple access requests, the unit with the highest priority is granted access
first. The hierarchy is established by simple field
engineering adjustments.
Modularity Both the memory size and the number
of input/output channels are expansible in small increments. For example, the memory of the B3500 is expansible from 10K to 150K characters; and the I/o
channels can range from 1 to 20.
Address Memory A special "address memory" is
used to increase the processing speed of the system. It
is located within the central processor. It consists of an
array of 24 words and is expansible in increments of 12
words to a maximum of 120 words. During execution,
the processor addresses core memory with words from
address memory, so that during the execution of an instruction, memory accesses are not required for infor-

mation relative to the command itself. The time required to access a word from address memory is only 150
nanoseconds.
Peripheral Equipment A wide range of conventional
peripheral equipment is offered with the B500. However,
only the Burroughs disk file offers clear competitive advantage over currently available peripheral equipment
from other manufacturers. The disk file provides online random access storage capacity of 10 million
eight-bit characters and is expandable to up to 2.5
billion characters. With the one-head-per-track construction Burroughs has cut down average access time
to only 20 milliseconds. This fast access time is coupled
with a relatively high transfer rate of 275 thousand
characters per second.
A new random access device called "system memory" is a single magnetic disk device capable of storing
up to two million characters of information. The primary
use of system memory is to house the software package
and the user program library. Average access time is
17 msec.
MagnetiC tape drives are provided with the system
both for seven and nine channel tapes with transfer
rates ranging from 36,000 to 144,000 characters per
second and recording densitites up to 1600 bits per inch.
Software Successful sale of third-generation computer systems implies good software support. Because
of the increasing shortage of competent personnel, computer users are becoming more heavily dependent on the
software support provided by the manufacturers. Software has been traditionally one of the weak spots in the
Burroughs product line; therefore it is with increased
interest that one examines the software available with
the B500 series.
The B500 offers both problem-oriented compiler
languages and machine-oriented assembler languages.
The problem-oriented languages include COBOL and
FORTRAN. Two versions of the machine-oriented assembler languages are available, an Advanced Assembler
and a Basic Assembler, both running under the control
of a Master Control Program. (Burroughs has not announced any plans to develop a PL/I compiler for the new
systems.) Thus, the software is heavily oriented toward
the financial, banking and retail community.
Program Compatibility The two systems are fully
program-compatible. The central processors also incorporate extremely fast read-only storage. This storage device is wired with interpretive routines which are
executed at a 50 nanoseconds cycle time. These routines
are called microprograms. The great advantage of
microprograms is that various different routines may
be wired in, allowing the "emulation" of other systems.
This same technique is used by IBM in achieving compatibility between System/360 and 1400/7000 Series
COMPUTERS and AUTOMATION for April, 1966

computers. Burroughs plans to use the emulation technique for both its own 200 and 300 Series systems and for
the IBM 1401 Series machines. As such, they will have
the only system offering a program conversion aid for
the over 500 current users of their 200 and 300 series
computers ••. thereby helping to insure customer loyalty.
Burroughs' Present Market Position During the last
decade Burroughs has gained a foothold in selective computer markets, such as banking, the military and the
government. Burroughs' current market position is the
result of concentrated selling efforts in these markets.
A breakdown by market category of the some $260 million
worth of Burroughs commercial EDP systems currently
installed is:
Finance, Insurance, Real Estate
33 %
Medical, Educational and Other Professional 16 %
Business and Other Non-Professional Service 13 %
Federal, State, City and Local Government
8%
Other
30 %
The selective penetration and comparative success which
Burroughs has achieved in these markets is due primarily to its line of peripheral equipment. The availability
of special peripherals, such as MICR sorter readers
and tape listers, are prerequisites for banking applications. Their two basic lines of central processors, the
B-5000 line and the B-200 line have not been overly
successful. However several times Burroughs has been
able to revamp its line by repackaging returned machines
and announcing them under new model numbers (e. g.,
returned B260/270/280's were introduced in 1964 under
the model numbers BI60/170/180). Also they have
offered various price reductions to encourage the sale of
their systems. These moves have effectively extended
the revenue-producing life of their computer equipment.
Burroughs' Expected Market Position Burroughs
expects its new systems to find wide use in on-line, realtime, and time-sharing operations such as automated
inventory control and distribution systems, on-line bank
systems, and management information systems.
Burroughs' President McDonald feels that while
large corporations, research institutes, and some
government agencies will continue to employ large scale
systems, the greatest growth potential for computer
usage in the next decade will be within the small to
medium size businesses and financial institutions.
"Inexpensive, automated, data capturing-andrecording devices, that prepare computer input should
make it possible for almost every business to afford the
control that accurate, timely, and properly organized
business data can provide." Despite this optimism, there
does not seem much reason to expect that Burroughs
will succeed in establishing a substantial market base
outside of ba.nKing and service bureaus. Burroughs can
continue to provide capable equipment for these special
markets, offering good system know-how and competitive
price/performance. By concentrating on these special
markets Burroughs should be able to maintain its current
overall market share of between 2.5-3.5 % in the years
ahead.

)

(

"-/

:;'Yj
'/ 11'-)

lU-

An announcement
of a new service

ffHE COMPUTER
DISPLAY REVIEW
Important advances are now being
made in man-machine commun ications, an area of computer technology
with far-reaching significance to businessmen, industrialists and scientists.
New .graphic and alphanumeric devices, connected to large time·shared
computers or small real-time processors. are already creating a major
breakthrough in this ar.ea. Display
devices "the windows of the computer" are making the computer
more useful to more people by providing instantaneous readout of data and
by accepting query or command inputs.
As pioneers and speCialists in .computer-based display equipment and
techniques, Adams Associates has recognized the need for a single source
of information and critical evaluation
of this entire field. THE COMPUTER
DISPLAY REVIEW, just published, answers this need. The information in it
results from an. intensive effort by
Adams Associates during the past six
months to gather, analyze and evaluate data on all equipment now available or under development in the free
world. By making the REVIEW available on a subscription basis to Corporate
Sponsors, its cost will be shared by interested firms and government agencies.
To keep sponsors current, the REVIEW
will be updated every four months. A
full-time staff will continue its research,
visiting equipment manufacturers and
field installations. New· developments
in display hardware, software, applications and trends will be thoroughly
evaluated and information on. them reo
leased. In addition, abstracts of timely
articles on applications and techniques
as we." as papers written expressly for
the REVIEW by well-known specialists
will be included in future issues.
For information on the content and cost
of the REVIEW, please call or write to
JohnT.Gilmore Jr., Vice President.
charles w.

adaIns associates
inc.
575 TECHNOLOGY SQUARE
CAMBRIDGE-MASSACHUSETTS 02139

COMPUTERS and AUTOMATION for April, 1966

(617) 491-6555

We will be very surprised if the rest of
the industry catches up with our new computer
system in the next 5 years.
It is called Sigma 7.
It does time sharing in a real-time environment.
It is a totally integrated hardware and
software package.
It was designed first of all to handle the
heaviest demands of priority-interrupt
processing - easily, simply, naturally.
Next it was designed to handle conversational
time sharing, and all the tasks of science,
engineering and business, concurrently, with
the same ease and naturalness.-

It can completely change its environment
from one PL/I program to another in 6
microseconds, under the control of an
operating system.
It can simultaneously perform real·time on-line
control, time-shared conversation, batch
processing, and high-speed input/output, with
full protection for every user.
The only system on the market now that
even·triesto do what Sigma 7 does costs six
times as much.
Deliveries will begin in six months.
With software.

Santa Monica, California
cities throughout the world

_<'':;~__,_~,_~_._,_j

Designat,e No. 9 on Readers Service Card

SCIENCE AND THE ADVANCED SOCIETY

C. P. Snow
Parliamentary Secretary
Ministry of Technology
British Government
London, England

"The computer
man."

the most remarkable machine by far) yet made by

The subject of my talk tonight is "Science and the Advanced Society." By "science" I think everyone knows what
I mean, and if you don't, I don't propose to define it
now. "The Advanced Society" however, I think I should
define. I mean by that, those parts of the world which have
solved the purely brutal predicaments of our fellow human
beings: societies which see that people get enough to eat,
get somewhere to live, get some sort of medical attention,
and can read. This, by our standard at the moment (and
it's a poor standard) is what I mean by an advanced society.
At the moment, advanced societies exist only in just over
one-third of the world: in the whole of North America; in
nearl y all of Europe (there are one or two pockets which I
think one could not honestly say were advanced); in the
Soviet Union; and in a few isolated enclaves in the rest of
the world, like Australia, New Zealand, and so on. These are
the advanced societies of our present world. They are doing
pretty well, and the reason why they are doing well is
perfectly simple. They are the countries which really did
take on what for want of a better term we call the Industrial Revolution. They were the people who learned to
make goods - through different histories, through different
processes - but they all learned very much the same techniques in the end.

Food and Transformations
With different degrees of success they learned perhaps
the most important trick of· all, which is to grow enough
food. Your country has been staggeringly successful in this
last task. At the moment I think I am right in saying that
6% of Americans can grow more than enough for the
whole of this gigantic country to eat and the real figure is
probably nearer 3%.
This has transformed your society. It has meant you
ceased to be an agricultural country long ago, because you
can do this trick so easily.
You're soon going to cease to be a manufacturing country. The great industries in America are going to be services
and education.
This has happened within a few years. To a lesser degree the same thing is true of all advanced societies. But it

means that we are going to have our hands full of problems.
I'm going to mention some of these in a moment or two,
but I'd like to say that almost all that I'm going to say
I've been discussing with my American colleagues like
Harrison Brown, Lee DuBridge, Glenn Seaborg - all these
excellent scientific administrators and the astonishing
thing is that we find ourselves completely in agreement.
We can see at least what the problems are; and it seems
to me, if we are in -agreement, it is more than likely that
there is something substantial in what we see.

The Industrial Revolution
The old Industrial Revolution was a messy affair. It
started in my country, more or less by chance. You took it
on very fast and fairly soon did it better. By about 1860 we
who had thirty years start or more were, in fact, not producing as much "as you were.
It worked. It worked in both countries. We also had certain advantages, both of us. It's a great advantage to have
a commercial civilization preceding an industrial one. It
teaches you to answer letters. It teaches you to do the
rudimentary administrative work which all industrial soci. eties require, although they can get by without it. So we
were in fact very lucky. The Russians were rather less lucky
because they had very little of this preliminary training in a
commercial civilization.
I think the Chinese in due course will make a very good
job of their Industrial Revolution simply because they
ryave a tradition of high bureaucracy. There has been a
very good Chinese civil service for about -two thousand
years, and two thousand years is a fair time. I would guess
they'll do it extremely efficiently, but, of course, they're
not there yet.
The Industrial Revolution had very little to do with
science. It was mainly the work of ingenious craftsmen,
people who could make things and didn't care much about
the ideas upon which those things were made. Men like
Henry Ford, who was not in any sense a scientist but had
a very shrewd idea that you could make a motor car. There
were many others, and almost the whole of at least the
first phase of the Industrial Revolution was done by
such persons.

The Scientific Revolution
(Based on a talk given at the Franklin Institute, Philadelphia, Pa., Jan\lary 27, 1965)

But now, we are entering something extremely different.
COMPUTERS and AUTOMATION for April, 1966

We are entering a revolution, which in the view of myself
and the friends I've just mentioned, 'is going to transcend
anything which we have so far known. The rate of change,
since about 1800, has been incomparably faster than anything men have known before. The rate of change between
1965 and 2000 will make the previous rate of change look
like a tea party. Of this there is no 'doubt. We must be
prepared for what this will do, both its desirable results,
and its undesirable results.
Let me be perfectly clear. I have no doubt whatever that
this revolution has to be done, and that by and large. mankind will immensely benefit by what has to be done. I am
not in the least pessimistic about the total social results of
what we're walking into. But it's as well to clear one's
head.
All great changes produce certain difficulties, and usually
certain losses as well as great gains. I think we're all fairly
sure, that the gains we're going to get are going to be
accompanied by certain losses.

The Computer Revolution
The great instrument, which is immediately about to
transform our lives, is the computer. The computer is not
just a calculating machine. It's a source of information, of
memory; an instrument which can collect information,
settle it, keep it, store it, analyze it, and so on. It is by far
the most remarkable kind of machine yet made by man.
I've heard, - this I don't believe, but I've heard it, some ingenious physicists say that by the year 2000 we shall
be able to make a computer which in every respect in imagination, creative power, and so on - is better than
any human brain yet existing. I don't believe this; but the
fact that the claim can be made by sensible scientists,
shows how remarkable these machines are, and, I think
it's fair to say, how little we really know about them. We
don't even understand yet their true nature.
We know that we can in theory make them so that they
can reproduce themselves. This I find a slightly creepy
thought. They're going to enter into the very texture of our
lives, as no machines have ever done before. Within a
short time none of you will ever write a check again.
Sometimes computers give rather surprising results, like
the case of a simple housewife in England, whose income
would be about $6,000 a year, who suddenly found from
her bank that her credit balance was 2,300,000 pounds! At
the moment computers can sometimes perform some rather
nonsensical operations. But in .fact computers are going to
do all kinds of things which we can hardly envisage. All our
payments, in and out, will be known. A great deal of the
details of our personal lives will be open and at the service
of a central government. They will make central government very much easier.

Submergence
This obviously has its disadvantages. The chief psychological disadvantage that I can see is that men may tend
to feel that they're being submerged by the technological
tide. We should be prepared for this. We should use every
resource of social compassion to make sure that this doesn't
happen.
In practi~al terms, a great many operations will be
done by computers which are now being done by men.
We've got to foresee this long before it happens. If we
have the foresight and the technological skills, most of those
practical effects - the non-psychological effects - can probably b.e at least mollified and, to some extent, wiped
away.
COMPUTERS and AUTOMATION for April, 1966

Non-Work
But there is a long-term effect for which I see very little
answer. It looks as though once you firmly establish a society
advanced in this cybernetic way, then ten or perhaps
twenty percent of the population will have to work extremely hard. Men of the whole managerial slice of society
will have more tasks, more difficult and more complicated,
than they've ever had before.
On the other hand, it seems to me inescapable that a
large slice of the population, perhaps as ·much as 80%, will
be underemployed, and will remain underemployed, so far
as anyone can see, forever. I think they needn't be underemployed for the next hundred years or so; but in the long
run, in a really advanced society, I can't see any easy way
out from this. People will in fact cease to have an ethic
based on work; they are bound to have an ethic based very
largely on non-work.
This is worrying me, but it has great advantages. If I had
to choose, I'd take it, because most men who ever lived, the
overwhelming majority of our fellow human beings, have in
fact lived lives which were short and hideous to contemplate;
they worked from the cradle to the grave, with extraordinarily little compensation. This has been the ordinary lot of
man from time immemorial. And so, therefore, to take
that burden away is fine, and I'll settle for that.

Purpose
But I think, again, we have to be realistic. We have to
remember that men and women want purpose in life. If
you take purpose away, then you're likely to leave a state of
boredom, ennui, a kind of feeling that society isn't theirs.
This, I believe, is ultimately .s0ing to be the real problem
for at least a certain part of the human race. They will feel,
"fine, but what's it all for? What is the point?"
Here, again, I think we should be preparing now for
this problem which is not immediate, but in this country
it is. not far away. If we get the rest of the world put to
rights, the problem must ultimately come to the rest of
the world also.
The absence of purpose is going to be one of the great
psychological chores, a much greater psychological chore
than anything that happened in the old Industrial Revolution
which made the shape of the United States and of most of
Europe.

The Biological Revolution
There is a second and probably a greater psychological
burden. It has not yet been carefully thought out, though
some of the problems of the cybernetic revolution have at
least been sketched by a :few of us. The biological revolution.
so far as I know, has only been announced, and its consequences not yet imagined.
Recently, I heard Charles Price, of the University of Pennsylvania, say· with his own authority, which is great, and
with the authority of a great many of the chemists of the
United States, that within a very short time we shall be
making living cells, quite simple livin~~J~~t, n~'yertheless,
Lord C. P. Snow, C.B.E., is the distinguished author
of "The Two Cultures" and m,any other books. He
is a scientist, educated at University College, Leicester, and Christ College] Cambridge, from, which. he
received his Ph.D. He has held positions for many
years in the government of Great Britain, often with
particular concern for selecting and placing scientists.

life in the full sense. I find this more remarkable than anything which the physicists and engineers have done, it is
something which is nearer the roots of our being, something
which will be extremely hard for many persons to accommodate to, just that fact alone, without any practical
applications.
There are practical applications also. As one of the two
prongs of this extraordinary biological revolution, we now understand a good deal about the mechanism of heredity. It
looks as if, not so immediately as the making of living cells, but
within foreseeable .time, we shall be able to control some of the
mechanism of heredity. This, again, is something so dramatic
that all the engineering triumphs of mankind will seem
comparatively pale beside it.
This is not science fiction. This is the considered view of
the best minds on this subject that I've been able to talk to.
Once again I suspect that the total result will increase
human good and not decrease it, although it is only fair to
report that one of the best biologists in the world, an
Australian called Macfarlane Burnet, said only a month ago,
that this was something man ought not even to try to know.
I can't agree with such a statement. I think that once you set
that kind of limit, that our lives will become impossible.
I think we have to try to know what we can know. But
I am sure that that somber warning ought to be borne in
mind by all of us, and ought to be borne in mind before
the thing happens.
Warnings

A great many things have taken human beings unawares.
There is no excuse whatever for these things to take us
unawares. We've had heaps of warning. We've got, with
the computers, two or three years before the full consequence
is upon us. With the biological revolution we've got perhaps
five to ten years before the secret is cracked. We ought to
be thinking now with all the imagination of humane persons
- sociologists, psychologists, anyone you like, any people of
good will who wish well rather than ill to their fellowmen.
This we've got to do; otherwise yo~'n get a great lack of
belief that there is any worth in .this life, among lots of
persons in the community who have talent but aren't robust
enough to face 'our life as it is going to develop.
There may be, as I say, an increasing lack of sense of
purpose: what intellectuals of your country are fond of
calling, alienation. I often think they call it alienation
because they would like to have been country squires with
large numbers of serfs, and they feel alienated from a
society which doesn't make it easy for them to have that
desirable result. They admire the peasant life having never
seen a peasant in .their lives. It would be very' salutary for
people who think of an 18th century society with happy,
jocular peasants, actually to go and see happy, jocular
peasants as they now exist in many places in this world.
Nevertheless, there may be an increase in the lack of human
purpose, a lack of the purpose which gives salt to life. I
suggest that for some time, for a hundred or two hundred
years, there is no excuse for people to feel that they lack
purpose. I've been talking exclusively for a while of advanced
societies, of the rich countries, of countries that have had the
historic luck.
The Poorer Countries

I want to draw your attention now to the rest of the world,
the poorer countri~s, the countries which have not had historic
luck; and so far as one can see, it is luck. There is very
little else that can explain what has happened to us.
The world is divided into rich countries (the ones I've

mentioned) and poor ones. The bitter fact is that the rich
countries are getting richer and the poor countries are
getting certainly relatively poorer and possibly absolutely
poorer; and no one can see yet how this gap can lJe
decreased. Curiously enough, to some extent within rich
~ocie~ies there is some sign of the same phenomenon; the gap
In thIS country, I suspect, between the rich and the poor is
not becoming any smaller. But that is a subsidiary problem
which can be coped with by intelligent political management.
The other problem, the gap between the rich and the poor
countries, seems to me by far the biggest social problem,
not only of our generation, but of at least three or four
genera tions to come.
It's extraordinarily hard for people born as lucky as you
are, living in this immensely rich society, whatever its
pockets, to understand even remotely what the life of twothirds of your fellowmen is like. A friend of mine sitting
next to me at an American dinner party in a very comfortable house said she had just been in an Asian town;
she was active like most Americans, and she saw people
lying absolutely still in the streets, not sleeping, not waking,
waiting for heaven knows what. And she asked, in a good
American way, "Why don't they move"? And her guide,
who was himself Asian, said "The less they move, the less
hungry they are."
Now this is so far from our experience, so far from anything we can even remotely come in contact with, that we
don't speak the same emotional language as the majority
of the human race.
'If you want to hear this very sharply from someone of
immense talent you should talk to one of the greatest of the
world's mathematical physicists, Abdus Salam, who was
born in a wretched Pakistani village as the son of a poor
villager. He often says, even to people whom he'd regard
to be of relatively good intentions, good heart, that: "Often
when you speak to me, you chill the blood in my veins. You
have no idea what our life is like and what the life of most
people is like and will continue to be for as long as we can
foresee."
I am thinking particularly of India, which is really the
problem that is weighing heavily on us all. There seems to
be no doubt that India is going to have a major famine
within a few weeks and several million people will starve to
death. Famine is a horrible thing.
I agree, of course, that there are certain exceptional
societies. In the Fiji Islands, you've got a very jolly people
who live comparatively simply, and live a fairly long time;
they are well-fed, and extremely athletic. They do nothing
but play games; I suppose this is fine. There are, of course,
other intermediate societies.
But, when you're trying to produce a sort of shorthand
account I think you've got to confine yourself to the gross
and major aspects.
Medicine, including surgery, spreads .much faster than
food growing and incomparably faster than industry. Therefore, the actual length of . life in countries like India has
gone up without any of the compensating things which we
had when our Industrial Revolution was upon us, without
being able to grow more food, make more goods, and so on.
Control of the population won't touch the \Situation during
this extremely critical fifteen years. The children are already
born, in fact. In all the poor countries of the world ~
fantastic proportion of the population, about 45%, is under
age fifteen. This is because in fact medical science goes
faster than either industry or food-growing. Therefore,
we've got to cope with a giant immediate problem.
The world could support a very much larger population
than it now has, probably three times as large or more. But
COMPUTERS and AUTOMATION for April, 1966

there is clearly a limit Almost all socIetIes in fact, even if
not informed, do limit their population. I do not believe
that overpopulation is as much a cardinal problem as some
of the others. I fancy that once we get a relatively well-fed
world, then you will find mysteriously that the population
will level off.

One-Third Rich, Two-Thirds Poor
I cannot believe that the world can survive in peace, onethird .rich and two-thirds poor. I simply do not believe it.
I believe that this tension is the profound conflict of these
days, and is· being reflected in the events of this tumultuous
century. People, once they get a little above the subsistence
level, want much more; and it is right that they should
want much more. I therefore believe, that all the rich
countries (chiefly this one, but all of Europe) will have to get
seriously to work if we want this world to be morally
tolerable or even, in my view, practically viable at all.
As I said, this country can contrive to grow enough food
for itself 'with probably the effort of two or three percent of
the population. Some of the best calculations suggest that
if we spend a few billion dollars, nothing like what we
spend on space or on war, we could get the agriculture of
these countries really going on their own terms. Then we
might help them keep up, keep up only this subsistence
level at which they are now existing. That is the best we
can do by any contribution in terms of skilled agriculturalists
and skilled equipment, and so on, for the poor of the world
for fifteen years.
Then, if we're going to make them slightly better nourished,
get them off the subsistence level, give them enough energy
ultimately to cope with their own problems, it will need a
contribution in actual kind, a contribution in terms of actual
food. Again, the amounts involved don't seem to be out-ofquestion large. It would mean paying the farmers. It would
mean some people ·working perhaps on the land who now
don't work on the land. But it looks like a realistic prospect,
that the 3% of your farmers who are now feeding the
United States could in fact grow a bit more without much
effort if they were paid for it. This, again, is not beyond
question. And we have made the same calculations for
Europe. It seems a perfectly feasible proposition.
If we do not cope with this task, then I'm afraid my
view of the world will become very dark. I cannot see any
conceivable solution for the world where we sit well-fed as
though we were in a kind of fortress, heavily armed, trying
to guard ourselves from the hordes outside. This is not
tolerable, at least not tolerable to me, and I can't live like that.
This is not a situation which a self-respecting human being
should be placed in.

whole of this planet, when in fact everybody is living at a
modest level, something like, say, North Italy today: not as
well as you (that's very difficult), but something which is
perfectly tolerable - North Italy or say Yugoslavia today.
\Vhat happens then?
Now, will all men, having solved all the gross problems,
all the problems we must solve if we're to think of ourselves
as human at all - when people are no longer hungry, no
longer short of medical attention, no longer seeing their
children die, no longer illiterate - will they then succumb
to boredom, ennui, all these things which in your literature
are so strongly represented?
Well, it may be, but I would doubt it. I think in fact
that men are much tougher than we think. The men of
the future won't have our problems; that's clear. They'll
have other problems. I believe in fact they will think our
fears were slightly absurd, though they will probably have
their own. There will be some of them who feel outside'
their society, because there have been some people in every
age who feel outside their society. But the better spirits, the
people who really know that man is a wild animal, that he is
at his best when pe is living in society - that he is a wild
and beastly animal often, but has certain capacities for
grace and certain aspirations - I believe they will say,
"After all, we did it. We made these machines. We've solved
these problems, and we're going to solve the others."
We came out of the caves, you know, about 12,000-13,000
years ago, perhaps. In the caves men painted pretty pictures.
They were people like us, they looked rather like us, but
life wasn't sweet. We've come a long way from the caves.
We shall go a long way further. The thing to do is not to
lose our nerve and to remember that we're all human.

The Effects of Science
But I'm on the whole optimistic. We've been talking
about science. Science is a future-directed activity. Optimism is in the very thread and cloth of science, because
science has always had its eye on the future, has always
progressed, and has always become better. Really, by and
large, despite all the despairing half-intellectuals, its effect
has been ultimately benevolent.
I believe we shall cope with this challenge. Challenge is a
thing meant to be picked up and coped with. Despair is a
vice and, hope, curiously enough, is a virtue. And we can do
something.

The Solving of Problems
I sometimes ask myself, though, what will happen when
we really have got some kind of social justice around the
COMPUTERS and AUTOMATION for April, 1966

See the DATA/620 at booth 1113 at the S.J.C.C.
Designate No. lOon Readers Service Card

..........•;.~.....~

COMPUTERS and AUTOMATION for April, 1966

to get data
fast enough for
timely decisions
hO\N

that they operate on the same permutation code (ASCII)
approved by the American Standards Association for
information interchange. This is the same language used
by many computers and other business machines.
Consequently, you can use Teletype sets to link your
branches and departments to a central real-time computer, feeding administrative data and engineering problems on-line into the computer from distant locations
almost simultaneously.
In addition, the Model 33 and 35 sets have typewriterlike 4-row keyboards that make them simple for any
typist to use.

to be made now-not later when information finally filters through the usual channels. Yet, how can you get
current sales and inventory information in time to set up
accurate production schedules or determine raw material
needs without committing more money than is necessary?
The answer is data communications.
Data When You Need It Most business information
needs can be solved easily with a data communications
system using Teletype terminal equipment-like Telespeed 750 (high-speed tape-to-tape) sets. This enables
you to get information where you need it, when you
need it. You are able to make better informed, more
timely decisions that could spell the difference between
profit and loss.
On-Line, Real-Time Operation A basic advantage
of Teletype Model 33, 35, and Telespeed equipment is

machines that make data move

Improves Decision-Making A major producer of
heating units uses Teletype Model 35 ASR (automatic
send-receive) sets to link distributing facilities in New
Jersey and Ohio directly with its home office computer
center. This company not only has cut as much as four
days off its order processing cycle, but also supplies its
management with up-to-date reports on company activities.

According to the marketing vice president, "this (system) enables better decision-making capabilities, permitting greater flexibility in dealing with customer demands."
Improves Management Control Data communications systems have helped solve many information problems, resulting in improved management efficiency and
control. That's why this Teletype equipment is made for
the Bell System and others who require dependable, low
cost communications.

For more about applications of Teletype equipment, write
for our new brochure, "WHAT DATA COMMUNICATIONS
CAN DO FOR YOU." Teletype Corporation, Dept. 88D,
5555 Touhy Avenue, Skokie, Illinois 60076.

TELETYPE

Designate No.

COMPUTERS and AUTOMATION for April, 1966

THE AUTOMATED PROCESSING OF PEOP,LE

Rod E. Packer~ Ph.D.
Senior Scientist
Dunlap and Associates~ Inc.
Darien~ Conn.

"The main train would never stop, thus solving a major accelerationdecele71ation problem in people processing, nearly nonexistent with
low-mass electrons in data processors."

Three years ago, a dozen of America's largest corporations
entered into unintentional competition to design and build
an automated people processor. Their prototypes - giant
architectural mechanisms - were assembled "at Flushing
Meadows, New York and tested almost continuously over
two years by the massive queuing of millions of people at
their input terminals.
Not too curiously, one of the slowest automated processors
of people was operated by the world's largest specialist in
automated processors of data, while one of the fastest movers
of people was built by the world's largest producer of human
transport vehicles. But the most efficient and effective people
-processor: of all was engineered by a concern experienced
in using technology to "simulate" the movements of people
in animated figures, images, and now, in .full-scale robots:
Walt Disney Enterprises. Even the impressive processor, GE's
"Progressland," with its respectable input rate of a person
per second, was unable· to prevent peak period queues of
thousands of people. But as sole survivor of the dismantling
of the World's Fair, it is the undisputed winner of the
competition, and will .continue in operation at Disneyland.
The "Progressland" processor's pre-programmed instructions perform an entertainment operation upon the people
being input, but its automated handling techniques are
those of modern data processors: simultaneous, electronically
synchronized routines, repeated upon "batches" of spectators
in six adjacent sectors of a revolving people-storage disk.
The I /0 conversions between sequential and parallel
handling of people (as individuals or as groups) is through input and-output buffers - moving ramps, escalators, delay
spirals and chambers, etc.

Just as the World's Fair's transportation exhibits were its
largest, so is the automobile industry our nation's largest.
The "motormania" of our auto-saturated populace (now
approaching one motor vehicle per capita) is, of course,
only our prevalent technology's implementation of a constant
characteristic of civilized humanity. We must move about.
We find it educational, essential, and entertaining. Only
recently have we discovered that most low-level "intellectual" activities consist mainly of moving data rapidly
and efficiently from one "address" to another. Complex
data-processing routines are built upon this fact. But long
ago we discovered, instinctively, that moving individuals
into new physical relationships with one another and with
their environment likewise initiates a wealth of higher-level
operations.
A very random, slow "influx" of people builds a "neighborhood," with its back-fence chit-chat. Faster, loosely controlled, movement creates and disperses "crowds," with
their simple emotional behavior. More selective transportation of individuals can convene an "audience" of similarly
inclined minds. Highly deliberate manipulation of physical
proximities within a large group will bring out either active
"factions" or passive "cross-section" committees. "Teamwork" - actual bodily coordination of the movements of
athletes or technicians - can produce results impossible for
an individual· alone. The forward movement of our society
is probably quite literally related to our ability to move our
society bodily - and will remain so until a future communications technology closes much of the practkal gap between
long-distance and face-to-face human interactions.

Merging Communication and Transportation
Automated Physical Transport of a Mass Population
Although uniquely spectacular, the moving "chairways,"
"people walls," "revolving auditoriums," and other
World's Fair devices for controlling the movements of masse's
of people were simply surface indications of the inevitable
application of automated process control to our basically
necessary systems for the physical transport of a mass population: to' rapid transit, air traffic, ocean and space-ship
navigation, and even to our hundred million' "personal"
vchicles that currently criss-cross each others' paths in
random abandon.

Such a "merger," in effect, of communications and transportation has captured the human imagination repeatedly.
Instant "transmission" of people from place to place, bit by
bit, was foretold in a "Lights Out" radio thriller in the B.C.
era (Before Computers). Unfortunately, in that tale, a
"programming" error reassembled the villain's mind in the
hero's body. The wheel's invention set civilization slowly
into motion, but we have since produced actual means of
transportation approaching our dreams of genie-power.ed
teleportation and Jules Verne trips around, into, and out
of this world. Our astronauts make it around the world in
COMPUTERS and AUTOMATION for April, 1966

80-plus minutes (and a programming error displaces splashdown by 100 miles instead of scrambling their brains).
Tomorrow's rapid-transit traveller is promised 400 mph
trains and 2000 mph jets.
The probably pointless shuffling of people, twice a day,
from here to there (and back again) is often attacked as
modern society's most conspicuous down-to-earth "accomplishment," and our crash space program its most costly
man-to-moon "lunacy." Unless more philosophic (and sedentary) minds can make us stop spinning our collective wheels
- rather unlikely - we will no doubt push on toward the
stars in multi-billion dollar spaceships that will, at last, let
us travel unendingly, throughout a lifetime.

Manual Control
Until now, virtually all travel has been individual; the
processing of people has been under the "manual control"
of each person. Departure times, direction, speed, and path
all have been largely uncoordinated. Full freedom of individual movement has been, in fact, so basic that detention
from it is our traditional social punishment. Our sprawling
network of highways alongside our withering passenger railroads paint only too clear a picture of obstinate determination of a hundred million individuals to preserve personal
freedom of movement - even if it means our rolling 3000
pounds of steel and chrome everywhere each of us goes.
But, as in the automation of less "personal' technological
systems, we must soon reluctantly relinquish manual control - even of our own bodies, at least in transit systems,1
to the automated mode: first, because the computer now
makes feasible automated passenger processing systems of
unprecedented safety, speed, economy and efficiency, and
second, because our existing transportation systems are simply collapsing, one by one, from age and overcapacity.

Its initial contributions, plus a switchover to high-capacity
jets, automation of flight-reservation systems, and nationwide construction of huge new airports (some with moving walkways, mobile boarding lounges, and even automated
departure displays) have relieved, at least temporarily, the
main pressures upon this particular form of people processing.
Long-range research has already produced completely automated take-off and landing devices (see Ref. 11} and concepts
such as integrated flight-path control from space satellites.
Local emergencies in mass car and rail transit are now
being compounded by related breakdowns in water and
power supply systems strained by the same explosion of
suburban populations. The early winter dusk demands of
uncoordinated commuter traffic on elevators, subways, and
street lights superimposed on normal lighting and heating
have been prime factors in creating conditions vulnerable to
"black-outs" in metropolitan London and the Northeast U.S.
The millions of autos have also been charged with air pollution, of course, and the near annihilaton of municipal
street transit and regional passenger rail systems.

Applied Research
Clearly, these crises now suggest serious applied research
into the automated movement of people. Half-hearted
preliminaries are past: the repainting of the same old
public transit vehicles in bright colors, "shopper buses"
and downtown "malls." Greyhound has pointedly ignored
commuters in its offer to "Leave the Driving to Us," leaving Hertz to "Put Yau in the Driver's Seat" for urban
driving. Desperation is now forcing the first steps, previously dismissed as radical or expensive, toward automated,
integrated city-transportation control. Naturally, initial efforts have dealt with automobiles, from which neither the
populace nor its politicians will be easily pried.

Street Traffic Bottlenecks

Flow Pattern

Street congestion pushed surface traffic onto "elevateds"
and into subways even before mass production began clogging city intersections with cars. Automatic signal lights
cleared these spot bottlenecks a bit, but created cross-town
obstacle courses. Expressways did away with the signal light
hurdles but produced blocks-wide jams at their flooded terminals, as downtown parking became impossible. Finally,
within the last decade, our explosive dispersal to the suburbs has caused the familiar daily commuting crush that
produces occasional 100-car collisions and regularly reduces
several cities' "free"-ways to impromptu parking lots.:!
Their municipal traffic systems h~ve, in effect, collapsed. A
bit belatedly, Detroit and Chicago have set up experiments
in expressway traffic flow control by electronic sensors and
automated entrance ramp feeder displays (see Ref. 8) .
Toronto is placing· its thousands of traffic signals under
centralized computer control. And San Francisco has begun a massive automated rail transit development to offs('t
peak-hour auto commuting (see Ref. 14).
Partial control systems had become, almost abruptly,
too meager for air traffic even sooner than for surface system's. In the late 1950's, Washington's municipal airport
had become so saturated under traditional coordination of
flight patterns that circling for landing often took longer
than the cross-country flight itself. Automated air-traffic
control (ATC ) has become a high priority research area.

The bare beginning, if people in cars are to be processed
like digital information in a computer, is to impose a basic
pattern upon the flow of vehicles along common paths and
through intersections. Such organization is introdued into
computing by central timing pulses. Toronto has already
centrally integrated the timing of a 1000-intersection signal
network through a Univac 1107. Sperry is designing for
Manhattan a $5.4 million electronic traffic "gating" system
in which 11 Univac 413 computers will coordinate flO\\,
through 1200 intersections, and through an additional 1500
outlying points on key feeder arteries. Sixty intersections
in San Jose, California are being placed under IBM 1710
control through feedback to its internal traffic model from
400 magnetic ~ensors buried at critical flow points.
Speed Control

lA briefly popular ballad of the' 1950's' related the plight of being lost
in the complexities of the MTA, Boston's mass transit system: ".
he never rrturned; his fate is still unlearned."

A second step is to automate traffic flow rate to'
equalize vehicle speeds. Besides increasing capacity, there
is a safety bonus in elimination of passing and laneswitching maneuvers. On a 3.2 mile test section of Lodge'
Dr. Packer's academic background is in human communications theory and mass education (B.A., Yale;
M.A., S.M.U.; Ph.D., Univ. of Minn.). His experience
includes television direction for both commercial and
educational stations and training systems engineering
for General Dynamics and Bunker-Ramo. The automation problems in mass transit are one facet of his
current interest in the human/computer interface in
systems designed for mass use.

20n December 30, 1963, Boston experienced a total traffic standstill,
city-wide, for five solid hours. (See "Getting to Work and Back," Ref. 6)

COMPUTERS and AUTOMATION for April, 1966

Expressway in Detroit, 21 overhead signs display speed
"recommendations" for each of three outgoing traffic lanes.
A human monitor of 14 closed-circuit TV cameras spaced
along the roadway can input 55, 40, or 25 mph displays at a central switcher. He can close any lane involved
in an accident by flashing a red "X" above it. Control
Data Corporation is automating this system, as an element
in the National Proving Ground for Freeway Surveillance
Control, by programming computer direction of new traffic
entry ramp displays.
Traffic Density Control

Control of traffic density, through input ramp closure is
being tried on four miles of Eisenhower Expressway in
Chicago, where automated density counts were begun in
late 1962. Feedback signals, now installed on west-bound
entry ramps, are metered to turn from green to yellow
when expressway density climbs near congestion rates, and
then to red when the ramp is closed to further traffic input.
A human operator must still activate the ramp closure
routine at a warning from the processor. This Chicago experimental hardware comes from General Signal, Automatic Signal, RCA, General Electric, and Honeywell.
Three-fourths of the project's $1.5 million funding is
Federal. With similar stimulus, automated traffic control
is being attempted in Atlanta, Cincinnati, Syracuse, and
Houston, and in England, Spain, Germany and Australia.
Computers will shortly control the basic auto traffic pulsing
of many cities.
Automation would next advance to the regulation of
spacing between vehicles, to direct road tracking, and finally
to programmed lane switching through vehicular requests
for entrance or exit. This means automated acceleration,
braking, steering and integrated entry/exit addressing. RCA
and General Motors have been collaborating for several
years, at David Sarnoff Research Center in Princeton, on an
experimental electronic system to implement such an automated expressway concept within 10 to 15 years. At an
added road-building cost of only 5% to 7% lane guidance
cables and detection loops would be buried in the roadway
and paired with electronic sensor/transmitter controls at
20-foot roadside intervals, dividing the road, in effect, into
digital blocks. The input to a control unit in each car from
the control units of successive blocks through which it
passes would regulate the vehicle in relation to feedback
from other automobiles in adjacent digital blocks. Roads to
be engineered to this concept, GM calls "Autolines."
Automated Roadway

Westinghouse's early research took a different slant. Its
"Roller Road," concept while accepting the near impossibility of separating people from their personal cars, circumvents the difficult interface between individually powered vehicles and the automated roadway by replacing the
20-foot electronic blocks with individually-powered rubber
rollers every 20 feet. Over these, flat-bottomed carriers
would ferry 10 cars each (packed sideways) along 3000mile, computer-controlled roller-ways at 150 mph between
major cities. Passengers would ride in a separate compartment and loading would be done hydraulically under computer control.
The compatibility of this trunkline concept with use of
100 million existing motor vehicles is foreshadowed in the
success of modules for freight processing being piggy-backed
on ships and rails. But the mass transport of bulky, empty
private cars simply for the convenience of their owners is
straining economics a bit, ignoring the universal availability

of rental cars (at ever-falling rates) and dismissing human
factors analysis. Human behavior is not that inflexible.
The first season of B&O rail-ferrying of family cars from
Chicago to Washington attracted only a few dozen drivers
away from turnpike tolls and fatigue. The multi-passenger
price of a rail ticket for their car deterred any mass
switchover to passenger piggy-backing.
Commercial truckers would undoubtedly hand manual
control of their vans over to an automated inter-city
truckway or rollerway (if their salaries were unaffected).
But the private driver will probably submit graciously in
the long run, not to a usurping of his car's controls by
automation,3 but to the sensible alternative of leaving behind, for pleasure driving, his ego-boosting powerbuggy,
and commuting or travelling "light" (i.e., his own compact
body only) for economy and ease of movement into and out
of congested civilization or over great distances. With this
in mind, researchers have turned their full attention recently to the automation of mass transit systems.
San Francisco Automated Transit

The politics of public versus private transit funding are
happily outside immediate discussion. San Francisco was
the first area to somehow have the motorists of its major
suburbs vote bonds for an automated comuter transit
system, to be in full 'operation by 1971. BART (Bay
Area Rapid Transit) construction will cost $1 billion. Additional R&D funds are from Washington, and operational
revenue will come from fares and subsidy from the Bay
Bridge's auto tolls. Details are impressive (See references
1, 12, 13 & 14), but the real significance is in the competition that will determine a practical overall system, already underway on the Diablo Test Track, 4.5 miles between Concord and Walnut Creek. Four computer-assisted systems are being tried: G.E., Westinghouse, General
Railway Signal, and Westinghouse Air Brake. The concepts
of the last two are advanced conventional track-circuit
controls. G.E. has designed a decentralized "guided radar"
system with small processors on -e~ch- train. Westinghouse
uses a central Prodac 500 computer, preprogrammed for
scheduling all trains, with real-time inputs from a Prodac
50 computer that monitors actual positions, speeds and ac··
celerations of each train via a "wiggly wire" along the
tracks. A control unit at each station transmits digital information conforming to a stored speed-distance profile for
that section of the track, to each train with the section. The
requirements for the acceptable automatic control system
are impressive: slowing, starting, door control and maintenance of train separations at 80 mph, and at 90 second
service intervals. Trains must stop within six inches of a
given spot.
In addition, automated fare collection is to be incorporated. Three teams have devised competitive hardware:
FMC Corp/Control Data will soon test theirs at two Cleveland transit stations. G.E. will test at its meter department
in New Hampshire; and Advanced Data Systems will further
test, in Los Angeles, a system it has worked with in London. One concept is of a cash-purchased multi-trip ticket
whose magnetically encoded value is decremented upon insertion in entry/exit pairs of gates. In the design of this
important interface between passenger and system, small
points are significant. Conveying the t~cket, while being
scanned, through the _~urnstile to allow passenger retrieval
3The usually sophisticated Cornell Aeronautical Labs recommended last
spring that any transition to "automatic guidance" of autos be attempted
first with mechanical devices - in deference to "a general lack of public
confidence in the reliability of electronic devices." Past inadequacies of
mechanical TV and data processing techniques should be enough warning
against the futility and danger of mechanical shuffling of people about
at high speeds.

COMPUTERS and AUTOMATION for April, 1966

~ ..

without breaking stride has helped prevent rush hour queues
in trials. Automated route displays are also being designed.
Should San Francisco succeed, several similar water-blocked
commuter areas are ready to follow: Philaelphia-Camden
and Washington, D.C. bottlenecks, for example, maximize
mass transit benefits. Los Angeles, nearing the end of its
freeway spree, would also like to turn to mass transit.
Public Acceptance and Use
The unmanned Grand Central Shuttle has been run in
New York since 1962, but neither it nor BART's advanced
technology are sufficien.t alone for mass transit success. The
sociology behind its public acceptance and use is crucial.
The automobile has been declared "still in the ascendancy"
in one official's opinion: "Every driver wants rapid transit
- because he hopes it will remove the driver in front of
him." Monorails, the early cure-all, lost favor because of
switching awkwardness and high initial cost. But automation of traditional subways has begun in Stockholm where
"automatic pilots" receive centrally integrated instructions.
And Montreal announced, in the fall, a $1.1 million contract
to Westinghouse Air Brake for automation of a 48-car "Expo
Express" capable of handling 30,000 visitors per hour to its
1967 World Exhibition.
Westinghouse is aggressively promoting an interim "skybus" expressway for urban public transport in Pittsburghsized cities. Unlike the earlier "elevateds," the electric buses
would roll quietly on pneumatic tires along relatively inexpensive concrete tracks. The system would ultimately be
under complete computer control: track-mounted communications wires would alow computer direction to be continuous, with emergency stopping if a vehicle goes 100 milliseconds without computer information (See reference 17).
M.I.T.'s exploratory study of "Northeast Corridor" transit
problems is being used by the new Department of Urban
Affairs to initiate major feasibility research on the 400 mph
rapid transit system between Boston and Washington.
The developmental engineering of a turbine-driven train
that "flies" through an underground tube (Ref. 7) is irresistible - as are exotic designs for VTOL .and hover
craft, a 700 passenger plane, an English Channel tunnel, a
2000 mph jet, etc. But each of these exotic engineering projects only adds to the dilemma of moving masses of people
from front door to millions of different destinations conveniently. A 400 mph train is unlikely to make many local
stops. Schemes are being devised for automated pick-up
and drop-off of "pods" at the front and rear of high speed
trains as they bypass parallel station strips. The main
train would never stop, thus solving a major accelerationdeceleration problem in people processing, nearly non-existent with low-mass electrons in data processors.
Collection and Dispersal
There remains the problem of efficient collection and dispersal at these local interfaces with the exotic system (See
ref. 16). Prediction is for greater use of small commuting/
shopping vehicles; first, special rental cars and "minibuses,"
then credit-card actuated "slot-machine autos," and finally
inexpensive electric "urbmobiles" rechargeable at parking
meters and so universally available as to be treated, within
a community, like shopping carts at present supermarket
plazas. A monthly utility charge would cover their use,
along with local power and water billings.
Increased attention to automating all transit facilities
may bring radical departures in design of "accessory"
systems: parking, elevators, escalators, walkways. Convenience to private auto~ has already changed hotels to motels, department stores' to shopping malls, parking lots to
high-rise automated garages. 4 Future elevator jescalator mergers may deliver a passenger, vertically and horizontally, to
a specific skyscraper office, just as data is addressed in a
COMPUTERS and AUTOMATION for April, 1966

core stack. Audiences may be guided to reserved seats by
an analog driven overhead "automated usher" whose spotlight pinpoints the seat whose number it scanned on a
magnetically-coded ticket stub. Moving walkways may funnel spectators into and out of jammed sports arenas and
subway stations at centrally integrated rates responsive to
the schedule of events or trains.
More IIUsable" Lifetime
As the human passenger is relieved of constant attention
to steering, spacing, switching, etc. - and as his trip becomes technologically smoother and less distracting, automation will hand us a bonus that sounds as remote to today's preoccupied participant in the commuter crush as jet
travel would have sounded to a Wells Fargo stage driver:
automated transportation will give us back a major fraction
of our usable lifetime, now wasted in just "getting about."
Radio has survived TV mainly by salvaging, for millions
of drivers chained to a steering wheel, some portion of a
daily hour or so otherwise lost entirely. Ford now offers
stereo tapedecks as optional equipment. Airlines serve meals
and show movies. Philadelphia is flirting with offering its
morning and evening rail commuters electric shavers and
plastic-packaged cocktails (respectively). Only a near-millionaire can now afford five days of relaxation aboard a
transatlantic liner in place of a jet flight whi.ch (with time
zone losses) doesn't even allow shirking of an hour of one's
work day. Automation, however, may provide each of us
brief intervals of in-transit time for socializing, reading and
recreation - and simultaneously eliminate the frenzied urge
we now feel to "hurry up and get there."
Data, which is ironically indifferent to delays, can be
processed in nanoseconds. But a shuttle service can never
be a shift register. The key to successful people processing
may, in fact, lie in tailoring the time for a trip to routine
human activities, not in speed and more speed. By creating
for each passenger a little world in which he can "accomplish," in transit, something personal and unrelated to the
process of travelling, the actual transit time becomes of
minor psychological importance. After all, when we come to
the end of a day well satisfied, it is with the living we
did in the preceding 24 hours, not with having traversed
some 20,000 miles in a circle to arrive back at where we
started.
Alice and the White Queen had to run just to stay in the
same spot, but people would very much prefer the re\ierse
- to sit comfortably still while moving rapidly forward.
Computers are already dispatching subway trains, manipulating auto registrations, and routing trucks and buses acros~
the country. It is not surprising that they will soon be
chauffering us around: perhaps not in a big black limousine with telephone and TV, nor in the past grandeur
of a private railway car, but at least in an efficient style that
will make travel once again a pleasure.

A Brief Bibliography from Recent Periodicals
on Automating the Moving of People
1.

"BART," Conway, Patricia L. Industrial Design, Oct.
1965, pp. 27-37. An illustrated description of the passenger design details in the prototype Bay Area Rapid
Transit System.
"Can Computers Call the Signals." Business Week, Nov.
20, 1965, pp. 80-91. Summary of automated traffic light
projects.

4Macy's has just built a circular multi-storied store in Queens where one
can park adjacent to any desired department sector via a continuously
enveloping ramp.

(Please turn to page 58 )

THE HUMANITIES AND THE COMPUTER:
SOME CURRENT RESEARCH PROBLEMS

Edmund A. Bowles
M anager Professional Activities
Dept. of Educational Affairs
IBM Corp.
Armonk N.Y. 10504
J

The reasons we are on a higher imaginative level than the Nineteenth
Century is not because we have finer imagination, but because we have
better instruments, which have put thought on a new level.
- Alfred North Whitehead

Within the past dozen years or so, the computer has made
itself felt in every aspect of our society. One hundred years
ago, it was the Industrial Revolution which wrought profound changes in the economic and social fabric of the
western world. Today there is an upheaval of comparable
force and significance in the so-called Computer Revolution.
Putting the steam engine on wheels is analogous to making
the computer accessible to a large number of people simultaneously at remote locations by means of typewriter consoles
and visual devices.

Principal" Advantages
Let us consider for a moment the principal advantages
of the computer to humanistic scholarship in general.
Its incredible speed allows a scholar to accomplish in a
short time what would otherwise take him a whole lifetime
of drudgery to accomplish. The computer shortens what
Ephraim Vogel has termed the "vision actuality interval"
which exists in every field of intellectual endeavor; that is,
the time-interval between a scholar's original idea or plan of
action and its ultimate fulfillment in tangible terms. In the
humanities, this time lag is distressingly long.
The computer's storage or memory constitutes an infinitely
more reliable respository than the mind of the proverbial
absent-minded professor. Indeed, once entries of any sort
are reduced to machine-readable form, the computer can
arrange . this information in any number of ways, finding
correlations that human brain-power would have missed.
The computer's great accuracy is completely dependable,
even when untold mountains of statistical data require
handling. Once the data has been checked for accuracy
before being placed in storage, it will remain so with total
recall.
And finally, the computer's automatic operation is not
subject to the vagaries of human fatigue, interruption, or
mood. Speed, memory and accuracy remain constant.

Saving of Time on Chores
To the humanist, I would suggest the most important of
these advantages is the immense saving of time gained by
the use of computers. It is useful to remind ourselves that
the Oxford English Dictionary took some 80 years to
complete with several generations of editors. Jakob and Wilhelm Grimms' monumental Deutsches Worterbuch began to
appear in 1854 and wasn't completed until 1960. Similarly,
the manual indexing of the complete works of Thomas
Aquinas (approximately 13 million words) would take 50
scholars 40 years to accomplish, but thanks to the computer,
the total time required by a few scholars working mainly
in Italy was less than one year. A concordance to the
Revised Standard Version of the Bihle was produced on a
high-speed computer within a period of several months;
this compares with the King James Version Concordance of
the last century which took 54 scholars 10 years to accomplish. The deciphering of the Mayan hieroglyphic
script by Russian mathematicians, we are told, took only 40
hours of computer time; human beings would have needed
thousands of man-years to accomplish this. All this l~ads to
the inevitable conclusion that there are a number of scholarly
tasks - call them the more tedious clerical chores, if you
will - that in this age demand the use of the computer.
Certainly, one can no longer think of concordances, dictionaries, or projects involving masses of statistical data and
numerous cross-correlations without bringing into play the
tools of data processing. Thus, the computer's power can be
harnessed to relieve scholars in the humanities of some of
their most burdensome activity while at the same time
providing their research with the benefits of greater speed
and accuracy.

Unsuspected Meanings
A second major advantage is that the computer is frrquently able to bring to light hitherto unsuspected relationCOMPUTERS and AUTOMATION for April, 1966

ships or meanings from the raw data on file. When descriptions of the orbits of 566 comets accurately tracked were
fed into a computer in order to test the generally accepted
theory that the orbits are randomly distributed, the results
showed that the orbits were clustered around a single axis
like the petals of an artichoke. Similarly, various puzzle
canons composed by Johann Sebastian Bach were fed into a
computer. In these usually two-voiced pieces, in which both
voices must have the same melody, Bach wrote out only one
voice, and the solver of the puzzle had to find the. spot
where the second voice enters, whether the time-value is
different, whether the original melody must be inverted or
retrograded, and whether it must start on the same note as
the original. The computer came up with the previous
solutions to the puzzle canons worked out 100 years ago - and
some new ones as well!

Suspicion, Fear, and Ignorance
Unfortunately there is a great deal of suspicion, fear, and
ignorance on the part of the humanist concerning the computer and its legitimate role in scholarship. This is no doubt
part of the larger view that an overweening science and
technology have conspired to form a monstrous ideology
that threatens man's human existence. Some see the machine
as eventually making decisions that man himself should make.
Others find sinister implications in every technological advance, maintaining the attitude that science and the humanities don't mix. Finally, there are those who, ignorant of
mathematics, fear they are totally and forever incapable of
comprehending the computer and therefore dismiss it.' Although no one would suggest burning at the stake the maker
of a computerized concordance, as was almost the fate of
the first person to make a complete concordance of the
English Bible in 1544, the computer-oriented humanist does
face some formidable opposition.
Ironically, while the impact of the computer may be compared to the influence of the Industrial Revolution one hundred years ago, there is also' an analogous reaction among
many highly-placed scholars to so-called computer-oriented
humanistic research. Not that any misguided intellectual
will physically attack "the dark Satanic mills," as did their
Luddite ancestors. Nevertheless, there are those who, knowing little of the computer's advantages of limitations, damn
the machine as not only useless but dangerous to the world
of scholarship, as their ancestors inveighed against the railroads with their coal, iron and steam. One conservative
scholar sent a computer-oriented friend a postcard with the
message, "On an IBM machine you commit greater ironies
than you discover!" Then there is the academician who
places no value on the scholar's time, who dismisses data
processing while pleading for more funds to allow for someone to be completely free for years in order to do the
necessary clerical work. One professor said that, "If you
have to use a computer to answer a question, it is not a
question which I would care to put." Fortunately for the
humanities, things are changing rapidly.
Let me now dismiss the negative position and turn to some
representative computer applications within various humanistic disciplines which demonstrate beyond any doubt the
role of this new device as an important and productive tool
of scholarship.

Archeological' Analysis
In the field of archeology, for example, the computer is
becoming invaluable in studies of shards or fragments of
artifacts found in the diggings of ruins. Two very basic
problems are classifying shards as to their cultural provenance
COMPUTERS and AUTOMATION for April, 1966

and reconstructing whole artifacts from broken fragments.
Jesse D. Jennings, an archeologist from the University of
Utah, has a collection of some 2,600 shards, each of which
has 50 attributes each. Obviously, this represents an astronomical number of comparisons to make by hand, and yet for
a computer it is a relatively simple task.
Similarly, Mr. Dee F. Green, Director of the University of
Missouri Museum, no doubt following the maxim that
pottery is "the alphabet of archeology," developed a code for
reducing the individual attributes of some 4000-odd pottery
vessels from burial lots in Arkansas to a numerical system for
computer handling. Once the material was classified, the
various attributes can be sorted into discrete categories and
then statistical techniques applied to lump the attributes
into statistically meaningful groups, or ceramic types.
As can readily be seen, such projects as these involving
many thousands of artifacts, each with numerous attributes,
as well as the dozens of correlations between them, really
demand the use of computers to handle the sheer mass of information and to derive really meaningful results therefrom.

History
Historians have been facing a new impetus for the application of social science research techniques to the analysis of
historical political data.
One of the earliest historical studies that used data processing was the study of Massachusetts shipping during the early
Colonial period made by Professor Bernard Bailyn of Harvard
University. Faced with the problem of sketching a realistic
picture of the subje-:t, he came upon a perfectly preserved
shipping record for an eighteen-year period containing information not only about the vessels registered but about
the owners as well; in other words, material of early American social and economic history. Bailyn not only summarized
and tabulated this data in comprehensive fashion but used
the opportunity to assess realistically the possibilities of
applying machine techniques to historical material and to
explore the problems of procedure.
The shipping register in question consisted of 1696 entries,
each giving information about a vessel and the people who
held shares in it. A total of 4725 punched cards were
produced which contained all the information available in
the register on the ships and their owners. Codes were
developed not only for the purely quantitative data but for
such qualitative information as names of people and places,
vessel types, occupations, building sites, etc. Although numerous problems were encountered along the way, it is
significant that Bailyn's book closes with the statement that
only with these tools and techniques was the analysis of the
register possible at all.
The American Historical Association has set up an ad hoc
Committee on the Collection of Basic Quantitative Data of
American Political History under the chairmanship of Professor Lee Benson. Election statistics on presidential campaigns from all counties in the United States from 1824 to
the present, roll-call votes during each Congressional session
since 1789, data on Federal court cases, and census and
ecological information are all being computerized. Further
material awaiting such attention exists in the fields of
Edmund A. Bowles is Manager, Professional Activities,
in . the department of Corporate Education of Inter-'
national Business Machines Corp. at Armonk, N.Y.
He was instrumental in arranging six regional conferences on the use of the computer in humanistic research, which took place at Boston Univ., Georgetown,
Princeton, Purdue, Rutgers, and the Univ. of Calif.
at Los Angeles.

agriculture, business statIstics, industry, religion, economic,
social, and cultural data, foreign trade, employment, tax
data, and housing. The amount of such unpublished information available is' staggering. For but one year in American
history, the U. S. Census Bureau Catalog includes over 5000
computer tape reels of data in the above-mentioned fields.
One wonders, indeed, how many other fertile fields of information await the computer-oriented humanist scholar!

Content Analysis
Another research project involves content analysis by means
of the computer. Professors Robert North and Ole Holsti of
Stanford University are analyzing the origins of World War
I by means of computer techniques for scanning and reporting the appearance of themes and relationships in a large
body of historical material pertaining to decision-making
during the 1914 crisis. "Communication is at the heart of
civilization," but since students of international relations are
considerably more restricted in access to data than most
social scientists - direct access to foreign policy leaders
is severely restricted - one method is to measure their
attitudes, values, and assessments by means of a computerproduced content analysis of political documents at times
of crisis. For carrying on further research, North and
Holsti have constructed a dictionary of 3521 critical words
and proper names, such as abdicate, abolish, accept, and
armaments, and Gromyko, Soviet Union and Communist.
These words are sought out in historical documents, and
changes in style are noted as the historical crisis grows in
severity in terms of verbal affectiveness, strength or weakness,
and activity or passivity. By this means we can explore,
for example, the relationship between the level of East-West
antagonism and the degree of cohesion between the Soviet
Union and China.

exciting promise for the future. The massive comparison of
text where there are several or even dozens of sources presents
an almost insurmountable problem for the scholar. To compare in complete detail as few as 40 manuscripts might
take the better part of a lifetime. It is this type of activity
that cries for the use of data processing techniques.
Perhaps ~he first such effort was the study by Professors
Mosteller and Wallace at Harvard University, and actually
continuing over a period of years, to solve the authorship
question of 12 disputed Federalist Papers. Briefly, literary
styles of Madison and Hamilton were identified, then matched
with the style of each of the disputed papers. Having found
such factors as sentence length, vocabulary and spelling
to be of no help (the two authors were remarkably alike),
it turned out that differences in the use of so-called key
function words particularly those of high frequency
such as trom, to, by, upon, also, and because - served to
pin down authorship of the papers in question. From
a number of computations, Mosteller and Wallace found that
most of the disputed documents were written by James
Madison.
Professor \Villiam Givson of N ew York University is directing "Project Occult," an acronym for Ordered Collation by
Computer of Unprepared Literary Texts. He and his collaborators wrote a workable computer program for the collation
of varying texts and the production of a readable printout. Using the final section of Henry James' short novel,
Daisy Miller in an early and late version (the latter heavily
revised), Flexowriter paper tapes of the two texts were
prepared. The computer program found and identified by
number the sentences which were the same or similar in both
versions. A print-out indicated either by code or actual
words or punctuation the degree of identity, sentence-bysentence.

Style Analysis
Word Indexes
Certainly the most well-known use of computers in the
field of literature is the construction of word indexes. Briefly,
there are two forms: first, a simple alphabetical listing of
text showing the frequency or location of the words or both;
and second, a textual concordance showing all the words of a
given literary work not only alphabetically but in context
as well. In this form, each word appears as many times
as there are words within the parameter arbitrarily chosen
for it along with the relevant passages of which it is a part.
Such a concordance is not only of immense value to a literary
scholar from the point of view of time saved, but is useful
also to those in other disciplines employing literature as
source material. For example, concordances to the poetry of
American authors have been issued since 1959 by the
Cornell University Press under the general editorship of
Prof. Stephen Parrish. In the case of the poems of William
Butler Yeats, an alphabetical index of all significant words
was constructed, each word shown in the line or lines of
verse in which it occurs. Opposite each line is given the
line number, abbreviated title of the poem, and the page
number on which the line occurs in the particular edition of
Yeats. An appendix to the concordance lists all indexed
words in order of frequency and gives their frequencies.
So-called concordance-generator programs are widespread.
Today, certainly, there is no justification for scholars spending years in manual indexing as was the fate of their
Victorian predecessors.

Textual Analysis
However, it is in the area of textual analysis that computer-oriented research in literature shows exceptional and

Mrs. Sally Sedelow of Saint Louis University has described
the use of the computer for a rigorous description and
analysis of pattern attributes of text. She makes the observation that while colleagues in linguistics have been making
major contributions to such fields as machine translations
and information retrieval - and in return, gaining important
insights into the structure of language - those in literature
have offered very little and gained very little. The aim of
such studies is to discover the differences between writers'
styles and to shed light on the changes of an indivMual
author's style over a period of time. Known as "computational stylistics," these techniques deal with the parameters
of literary style in terms of its constituent elements: rhythm,
texture, and form.

Musical Structure
Musicologists, while perhaps overshadowed by their musician-colleagues who let the computer "compose" musiC or
use it to fashion new sounds, have nevertheless been very
active at the forefront of scholarly data processing. Over
fifteen years ago, for example, Professor Bertram H. Bronson
at Berkeley used computer techniques in an analysis of the
structure of folk-songs; By means of punched cards, he
coded the important elements of folk-tunes, including range,
modal characteristcs, prevailing time-signature, number of
phrases, the nature or pattern of refrains, final cadences,
and so forth. By this means, an entire corpus of 'folk song
material can be recorded both fully and accurately. The
various elements can then be analyzed for statistical patterns,
comparisons, or indeed subje~ted to any other query consistent with the data.
COMPUTERS and AUTOMATION for April, 1966

A computer can also serve to test hypotheses through
simulation and models. Dr. Allen Forte of Yale University
is applying machine analysis to help provide insights regarding the structure of the atonal music of Arnold Schonberg. The anatomy of a so-called pre-twelve-tone (or nontonal) music is still somewhat of a mystery. Mr. Forte has
found the traditional non-machine forms of analyses lacking
and has stated that a structural description of this music
would be virtually impossible without the aid of a computer.
If I understand the outline of his program, first, he is
formulating a basic working theoretical hypothesis based
upon linguistic and mathematical models for musical structure; second, he is developing an analytical method to
explore his ideas by means of the computer; and finally he
will test the result. Harry B. Lincoln of Harpur College is
using the techn:ques of information retrieval to compile a
catalog of musical incipits (that is, brief melodic quotations
of the first six to eight notes of a composition), of the
entire body of 16th Century Italian frottole, a known body
of some 600 polyphonic compositions. Since the possible
permutations and combinations of the beginning notes of
such pieces are almost infinite, both as to pitch and rhythm,
such brief quotations represent unique identifications of the
compositions from which they are taken. First the incipit is
translated into alpha-numerical form by means of a code
which can activate a photon printer. This is a device ·with a
high-speed rotating disk containing about 1400 characters
(in this case, musical), a light source, lens, and photographic
film. This code consists of .even numbers for the spaces,
odd numbers for the lines of the musical stave, an H for
a half note, a W for a quarter note, and so forth. One
punched aperture card with a 35 mm photograph of the
particular score set in the right-hand side holds information
such as the composer, title, voice part, accession or serial
number. The second card contains the proper sequence of
notes representing the incipit coded for the photon device.
When computerized, this code causes the printer to raise or
lower its focus to the proper line or space and, when th~
correct note or other symbol is in place, shoot a beam of
light through the proper aperture in the disk exposing the
film at that time with the desired musical symbol. At the
same time, a computer program extracts from this coded
information the intervalic order or melodic profile of the
particular incipit. This, then, can be compared to other
musical sources for instances of borrowings by one composer
from another, or from other works of the composer himself.
All this suggests that new technologies, coupled with computer-oriented research, will open up an added dimension
to humanistic scholarship.
Immediate Needs
What are the immediate needs?
• Courses in computers and programming tailored to the
specific requirements of the humanities and using examples from these various disciplines;
• A manual or textbook explaining the achievements
and potentialities of data processing in the humanities,
presenting in non-technical language the fundamentals
of computer science, and defining the scholarly tasks
which can be greatly speeded up by computers;
• A professional society with its own journal devoted to
computer-oriented research and to disseminating information about current projects and special computer programs;
• Creative computer-oriented professors on the faculty of
every la·rge humanities department and humanisticallyoriented personnel in university computer centers; and
finally,
• Fellowships and scholarships for dissertations and' research involving the use of the computer as a tool.
COMPUTERS and AUTOMATION for April, 1966

The Information Explosion
One final problem: How can we handle the "information
explosion" which is spreading rapidly from the sciences to
the humanities? Knowledge is doubling every ten years, and
scholarly journals are proliferating at the rate of over three
per day. How can we integrate and disseminate this continually growing mass of information? Following the lead of
the social sciences and their National Council on Social
Science Data Archives, the humanities sorely needs a planning,
policy-making and information-disseminating organization for
establishing, maintaining and coordinating data archives in
the several disciplines. Who is aware, as of this moment, of
all the computer programs useful in humanistic research,
or indeed where tapes are available containing masses o.f data
or texts in machine-readable form? The Library of Congress,
for example, set up a National Referral Center for Science
and Technology which collects data on information sources
and makes it available to scientists. In addition, the Smithsonian Institution has a Science Information Exchange
which receives notices of current research projects from
scholars. About 70,000 notices arrive each year and are
stored on computer tapes for retrieval.
. It is worth noting that 95 per cent of inqmrmg scientists
knew nothing of investigators working elsewhere on the
same problem.
The Computer and the Humanities
In a lecture at M.LT., entitled, "The Computer in the
University," the prediction was made that in a few years the
computer will have settled immutably into our thought as an
absolutely essential part of any university program in the
physical, psychological, and economic sciences. On the
basis of what I have said, I think the time has come to
amend that statement to include the humanities. Furthermore, within a short time, I believe a knowledge of data
processing will become part of the "common baggage" of
research tools and techniques required of every graduate
student in the liberal arts. I am even tempted to go a step
further and state that with the increasing number of courses
in programming being offered at our universities the time
may come when some students in the humanities may be as
fluent in programming as in writing English composition.
Certainly, the computer is fast becoming an important and
indispensable research tool for faculty and students alike.
The value of such an acquaintanceship can be seen in the
case of a professor of art history and archeology at an
eastern college. Describing himself as probably the man on
campus "least likely to benefit from a computer," he took
a short summer course at the college computer center.
Later, in reporting on the instruction, he said:
"The course profoundly affected the thinking of all of us.
This is the important thing - much more important
than the machine itself. Of course, we know that it is
the brains behind the machine that make. these miracles
possible. Nonetheless, it is a weapon of ~uch power that
all intelligent men and women everywhere should know
the kind of things it can do. Once we know that, we
can devise ways to make use of it.
Let us, therefore, see the computer as a means of liberation, freeing the humanist scholar -from the time-consuming
operations of the past, a tool rapidly providing him with
proliferating resources in the form of statistics, collations,
print-outs, cross-references, frequency-counts and hypothetical
models upon which he may build a research of new dimensions and complexity. Viewed in this light, it is a device
the potentialities and applications of which we cannot
afford to ignore.

Introducing two new
Burroughs 500 Systems:
B 2500 and B 3500
t

They're built to respond to ~our needs
instead of making you respond to theirs.
These two new user-oriented
computers are the latest Burroughs
500 Systems to be built by teams of
hardware and software experts.
Burroughs started this new
trend in 1960 with the B 5000,
which established the value of integrated hardware/software design.
28

That system's more powerful successor, the B 5500, is still unmatched
by the new competitive systems
that are modeled after it.
The other Burroughs 500 System, the B 8500, is the most powerful computer system yet designed.
Now, this new level of computer

responsiveness to business and
scientific problems is available to
even the smallest organization with
a requirement for electronic data
processIng.
Here are just a few of the impressive characteristics of the new
Burroughs B 2500 and B 3500:

COMPUTERS and AUTOMATION for April, 1966

1 Extremely fast hardware improve supervision by facilitating
speeds, with multimillion digit-per- review and control of programs and
second data transfers and control by demanding standardized documemories that operate in billionths mentation. They improve commuof a second.
nication by removing the "machine
2 The ability to do many un- language curtain" between those
related jobs at once (multiprocess- who understand the problem and
ing)-and to continue doing them those who understand the computer.
6 A special suitability to
without interruption even if
you drop in a rush job on
real time, data communicathe spur of the moment.
tions and time sharing prob3 An unprecedented
lems. With the B 3500, it is
possible, for example, to
degree of self-regulation in
low-cost computer systems
process order-entry from
via a choice of two operating
remote locations and compile
systems: the Basic and
from remote locations and
Master Control Proexecute major production runs at the comgrams. The Master
puter site-all at once.
Control Program, for
example, not only
Yet every program is
written solely to solve the
does more, but requires far
less resident core memory
problem it was assigned to
than any other on the marhandle. Automatic interrupt,
keto It provides automatic
full memory protection, an
scheduling, control over
interval timer, program segmultiprocessing, memory
.mentation, automatic priority scheduling and other
and I/O allocation, automaCentral
features combine to provide
tic maintenance of a library
Processor
quick response to a wide variof programs and data, program selection and initiation, error ety of simultaneous demands, with
correction functions, interrupt no interference between jobs.
handling, maintenance of the
7 The ability to accommodate
the fastest random access disk file
system log, and much more.
4 Programing that's so simple on the market. Operating speed of
it can be started
this already fast
by one programdevice may now be
multiplied by simulmer and con tintaneous use through
ued by another
-or divided up
up to four I/O
and then inteChannels.
gra ted by the
S From 4 to 20
operating sysI/O channels (all of
which may be active
tern. Since the
housekeeping
simultaneously and
details are taken
still leave ample time
free for computaover by the operating system,
tion) plus mUltiplexthe programmer
ers and exchanges
is free to conthat allow great flexibility and simultaneity
c e n t rat eon
the problem,
of I/O operations.
not 0 nth e
9 Monolithic integrated
machine.
circuit design which produces
5 Higher level programming greater speed and reliability and
languages (COBOL and Fortran) reduces size and costs. The B 2500
which save time and money. They and B 3500 make use of two proven
COMPUTERS and AUTOMATION for April, 1966

concepts in the very forefront of
this development: complementary
transistor logic and array monolithics.
10 Emulators which make it
simple and quick to convert from
our B 200/B 300 systems or from
1401, 1440, or 1460 systems. These
conversion aids make your old programs immediately usable on the
faster, more powerful B 2500 and
B 3500.
11 A responsiveness to change
in all aspects of computer usefrom a change in the number of
peripheral units to a change in program priorities. From a switch of
card to tape or from random access
to real time. And under MCP control, when you add new components,

more memory, more I/O capacity,
or upgrade from a B 2500 to a B
3500-anything short of changing
the basic method of processing data
-absolutely no reprograrning is nee-:essary. Change on these systems is
economical, quick, and orderly.
The B 2500 and B 3500 share in
one other important characteristic.
They are both products of Burroughs acknowledged excellence in
electronic data processing.

Burroughs
Corporation Burruui
Detroit, Michigan

Circle No. 12 on Readers Service Card

48232

c&a
THROUGHPUT

The Tower of Babel Revisited
I t is high time that data processing terminology and argot
be refined, defined, and straightened out. It is somewhat
frightening to notice the many terms and acronyms which
are used today without uniformity and with confusion. In
addition, the number of new terms which the industry sprouts
appears to rival the number of new computer installations some 6,000 per year.
If we try to categorize this confusion, we find it in three
important areas:
1-

In basic terms, used in day-to-day communication
among data processing personnel,
2 - In advanced terms, used among the elite few, who
may be embarrassed to admit their lack of understanding,
3 - In acronyms, used to abbreviate both of the above,
which replace confusion with lack of understanding.
Consider some of the basic terms. One of the most obvious examples is the name of the industry. "Data Processing"
is generic, all-encompassing; "Automatic Data Processing"
(ADP) survives mainly in Washington; "Electronic Data
Processing" (EDP) may be used to describe an equipment
manufacturer's representative (as in "EDP salesman"); "Information Processing" is more sophisticated; "Information
Handling" is broader; "Automation" is not acceptable to
organized labor.
More confusion is apparent among basic technical terms,
where definition is more important. Consider the process
which describes pictorially the flow of a system, or the flow
of a program. For this, a· profusion of terms exists, each
almost completely interchangeable with the other:
block diagram
flow chart
process chart
flow diagram
block chart
logic chart
logical flow chart
program chart
system chart

Modified
By

detailed
semi-detailed
macro
micro
"big picture"
program
systems
overall

hlock diagram
flow chart
process chart
macro
micro

to depict the logic of a program
to show the flow of a system
to show the flow of a series of manual processes
to indicate the total program or total system
to indicate an element of the total

Confusion among the "advanced" terms is also obvious,
and discouraging to persons newly in data processing. Consider the following terms, and try to define them, if you can:
real-time
on-line
multi-programming
multi-processing
time-sharing
software
These are popular terms; yet there is no uniform understanding of their meaning. If, for example, two programs
are in memory simultaneously (presumably multi-programming, but often confused with multi-processing) and one is.
on-line on a real-time basis, and the other is a time-sharing
system, what is the total system called? I suppose the
answer is software.
Some acronyms which pun, confuse, or distract at the
expense of meaning include:
BEST (NCR Business Equipment Systems Technique);
ARE (Advanced Real-Time Executive);
BASIC (Beginners' All-Purpose Symbolic Instruction
Code);
TO (Trade-Off);
OF (Old-Fashioned);
MAD (Michigan Algorithmic Dialect);
THUGS (Two Hundred eser Group, Southpast).
One of the tasks which joint computer conferences, data
processing management meetings, and similar associations
and councils could well take on, is cutting down the rank
growth of these weeds. Well-selected terms clearly defined
and some informal joint promotion of their use might make
a big contrihution to understanding and education.

The U. S. Government Glossary, published by the Bureau
of the Budget, does not help, since all of these terms appear.
(Since this Glossary spells programming and diagramming
with a single 'm', it may not become the national authority.)
If a choice for a standard here is desired, the following
is suggested:

Dick ·R. Brandon
Contributing Editor
COMPUTERS and AUTOMATION for April, 1966
Readers .Service Card No. 13 ~

HoneYlA(el1
rep'orton
Data Communication

One ola series on topics of importance to data processing management

Managements across the country are combining the talents of the computer with modern communication techniques to achieve more efficient operations, improved use of corporate resources, tighter
control and coordination of operating elements, and faster response to transactions. However, these
benefits, and the potential cost savings associated with them, can come only to those managements
t~at have c'arefully evaluated the response requirements of their data processing operations and have
made these major considerations in the choice of a data communication system. This report indicates
opportunities for effective and economical data communication, discusses requirements for an "ontime" system, and summarizes the equipment and services available to Honeywell Series 200 users.

HONEYWELL REPORT ON OAT A COMMUNICATION

DATA COMMUNICATION ADDS REACH TO
YOUR COMPUTER
The ability of a data communication system to link computers, or to extend the power of a central computer to
remote locations, has tremendous potential for business, industry and government. More specifically, some
. of the basic functions or activities that are benefitting
from this capability are:

Customer Order Entry - where the nature of the
product or the market necessitates immediate
response as to order status.
Control of Irreversible Transactions - where
complex analyses of information to support such
actions as the granting of loans and credit are
required for management decisions.
Data Collection - where volume, tendency for
human error, and extent of processing require
greater discipline in collection methods.
Control of Interdependent Operations - where
involved and complex operations such as job
scheduling and production control can benefit
from faster exchange and processing of data.
File Interrogation - where dynamically changing
information such as stock price quotations must
be constantly available for quick decision.
Customer Service - where response to customer
queries within limited waiting time is desired, as
in a hospital admission or savings bank system.
Information Retrieval - where provision for highspeed insertion, deletion, and access to large
volumes of textual material is a requirement.
Although each of these application areas is distinguished by its particular response requirements, there is an lion-time" attribute common to
all. "On-time" in one instance may mean instantaneously, or in real time. In another case
"on-time" might well be within an hour, a day, or
even a week. Since the cost of a system increases
rapidly as the response time of the system
decreases, the "on-time" requirement of an application becomes an important economic consideration.

THE ON-TIME SYSTEM
A significant feature of the communication-based ontime system is that it places a heavy demand on the
computer manufacturer to provide system elements
which can function in a wide range of on-time situations.
The following sampling of Honeywell data communication applications indicates the flexibility that can be
achieved when a product line is geared to this design goal.·

A large distributor handles 3,000 orders per day on
an inventory of 20,000 items by linking two Honeywell
computers at the home office to several warehouses via
teletypewriter. Upon receipt of a warehouse order, the
computer runs a credit check, computes quantity, brand,
size, and price, and transmits the totalled invoice back
to the warehouse in only minutes.

A large manufacturer uses a Honeywell computer to
control message switching for a nation-wide network of
100 teletypewriter stations concurrently with data processing. The computer receives the message from the
sending station, stores it, and forwards it to the receiving
station upon availability of an outgoing line.

A racing association system uses two Honeywell
computers which, in conjunction with ticket-issuing
machines, record all types of pari-mutuel bets and compute odds and payoffs instantaneously.

A Honeywell system handles some 300,000 inquiries
per day for a national credit bureau. All credit inquiries
are answered within 24 hours.

A trucking firm uses a Honeywell data communication system in which several freight terminals can be
linked with the home office. Among other things, the
computer calculates charges and transmits final freight
bills to the destination terminal before the arrival of
the trucks.

A telephone company uses a matched pair of
Honeywell computers to provide long-distance operators with split-second voice response to their queries on
rate information. The system handles 5,000 inquiries per
hour from operators throughout a five-state region.
Formerly, it took operators using a rate book 45 seconds
or longer to determine the rates.

HARNESSING TWO TECHNOLOGIES
The foregoing examples illustrate the diversity of applications and organizations now using data communication. This diversity will multiply in the near future as data
communication developments continue to occur at a
rapid pace. Already, central processors have made significant advances in their ability to control large-scale
inquiry, data collection and message-switching systems. A greater range of more economical and sophisticated terminals is appearing in the marketplace.
Systems design is maturing as evidenced, for example,
by more efficient joint voice-data use of telephone services. Communications facilities, services and tariffs
offer more flexibility than ever before.

Since the computer is the hub of the data communication system, it is up to the computer manufacturer to
provide facilities that will fully exploit the systems design
flexibility offered by proliferating developments in communications technology.

per second and a total of 7,000 characters per second for

all lines, a rate exceeding the requirements even of highvolume, 63-line message-switching systems.

DIMENSIONAL DATA COMMUNICATION
Dimensional data communication is one facet of the
"dimensional data processing" concept underlying the
design of Honeywell Series 200 systems. Under this
concept, Series 200 capabilities are available in small
increments making it possible to tailor a Series 200 system to meet both the functional and capacity requirements of a user's job. He needn't be saddled with oversized and costly capabilities which he does not need.
Furthermore as his workload increases, capabilities can
be added or expanded, gradually and economically. Inasmuch as data communications involve both a computer and communications facilities, here's how the
dimensional concept applies to both of these.

In addition to the wide range of non-Honeywell devices
that it can accommodate, Honeywell's Series 200 includes its own excellent terminals. These include CRT
display devices and the Data Station, a multipurpose
remote terminal. The Data Station features several optional capabilities including direct keyboarding, printing,
card reading, paper tape reading and punching, and
optical bar code reading - a unique capability for online or off-line handling of returnable documents such
as insurance premiums and utility bills.

A Partial Listing of Communication FaCilities That Can
be Incorporated in a Honeywell Series 200 System
Terminal

5
5

e
e
It
If

n
!r
I.
S

f\
Y

iL-

a
a
I-

e
;;3.
::lI

Service & Line

Data Set

Transmission
Speed

COMMUNICATION FACILITIES

DATASPEED I 2

Voice-Grade Private Line
DOD

202D
202C

105 cps

Most of the elements in a communication system such as terminals, data sets, and communication lines
- are available in a wide variety of types and capabilities
and thus offer ample flexibility for precisely tailored systems, initially and as the user grows. However, in order
for a computer to communicate over a particular line and
with a particular terminal, the computer manufacturer
must provide a communication interface designed to
handle that specific line-terminal combination. The
interface is therefore the key to flexibility and the cornerstone to economical systems design. The greater the
variety of communication facilities that the computer
can handle, the greater the chances for a system of
optimum design.

DATASPEED I 5 RECEIVERS

Voice-Grade Private Line
DOD

402C

75 cps

DATASPEEDI 5 SEND UNITS

Voice-Grade Private Line
DOD

4020

75 cps

DIGITRONICS
DIAL-O-VERTERZ

Voice-Grade Private Line
DOD

2020
202C

150 cps

DIGITRONICS TYPE 1
DIAL-O-VERHR2

Voice-Grade Private Line
DOD

. 201B
20lA

300 cps
250 cps

FRIDEN COLLECTADATA3 30

Voice-Grade Private Line
DOD

103F
103A

30 cps

Honeywell Series 200
Computer4

Voice-Grade Private Line
DOD

201B
201A

300 cps
250 cps

Telpak A
48 KC Broad-Band Channel

301B

5100 cps

Honeywell Data Station

Voice-Grade Private Line
DOD

2020
202C

120 cps

IBM 1050

W. U. 180 Baud
Tel. Co. 150 Baud
Voice-Grade Private Line

1181.1A
816
103F

14.8 cps

Tel. Co. TWX-CE
Tel. Co. DOD

103A
103A

14.8 cps

Voice-Grade Private Line
DOD

2020
202C

150 cps

Voice-Grade Private Line
• DOD

201B
20lA

300 cps
250 cps

Honeywell Series 200 systems are available with a full
range of communication interfaces providing an extremely broad selection of line-terminal combinations
(see accompanying table). Furthermore, the number as
well as the variety of lines and terminals that can be
combined in a single system are sufficient to fill the
requirements of any application.
Tho Series 200 interface capability includes both singleline and multiline communication controls. Both controls are available with either character-by-character or
message modes of operation. The single-line controls
can send and receive data at the high speeds available
through TELPAK, and still faster units can be provided
on special order. The multiline controls can handle
transmissions over as many as 63 lines simultaneously.
It can accept varied combinations of remote terminals
and can handle lines with speeds up to 300 characters

IBM Standard STR Series
(7702. 1013, 1009, etc.)

TTY 15, 19,28

5-Level TTY Circuit

60,66,75,
or 100 wpm

TTY 33,

TWX
TWX·CE
Tel. Co. 150 Baud
DOD

811B
103A
816
103A

100 wpm

TTY 33, 35, 37 Modell

Voice-Grade Private Line
W. U. 180 Baud

103F
1181.1 A

100 wpm

UNIVAC 1004/DLT2

Voice-Grade Private Line
DOD

201B
20lA

300 cps
250 cps

Telpak A
48 KC Broad-Band Channel

301B

5100 cps

W. U. Telex

W. U.
Adapter

35~

. UNIVAC 1004/DLT2B
W.U. TELEX

ITrademark of American Telephone and Telegraph Co.
2Trademark of Digitronics Corp.
3Trademark of Friden, fnc.
4This capability handled by single-line control only
5This capability handled by multiline control only

66 wpm

HONEYWELL REPORT ON DATA COMMUNICATION

THE COMPUTER
All Series 200 processors are program compatible. This
and the fact that Series 200 encompasses a variety of
input/output devices, all of which are available in many
levels of capability, enable the user to tailor his system
to the exact dimensions of his data communication job.
As an example of the Series 200's modularity, memory
for the Model 200 starts at 4K characters and can be
enlarged in 4K increments up to 32K and, from there, in
increments of 8K up to 65K characters. Similarly, memory cycle times in the Series 200 start at 3 microseconds
per character, then drop to 2, 1.5, 1, 188 nanoseconds,
and 94 nanoseconds. Hence, the appropriate processor
speed and memory size can be selected to meet both
conventional and communication loads.
A unique input/output scheme and a hardware interrupt
capability enable Series 200 systems to handle communications while simultaneously providing high production rates on conventional data processing applications. All Series 200 systems can handle multiple input/
output data transfers simultaneously with computing;
thus, they can send and receive messages over communication lines at the same time that input/output devices engaged in regular production runs are running at
high speeds.
Series 200 also offers a wide range of peripheral capabilities for both real time and batched processing needs.
There are 13 magnetic tape units ranging in data transfer
rates from 7,200 characters per second to 96,000 characters per second. Honeywell's new Mass Memory File is
available in three models, offering a range of on-line
storage capacities up to 2.4 billion characters per control
unit and random access-times as low as 95 milliseconds.
For faster access when storage requirements are less,
a control/drum subsystem holding up to 20-million
characters provides access to any record in an average
time of 27.5 milliseconds.

TO: Honeywell EDP
60 Walnut Street
Wellesley Hills, Mass. 02181
Attention: Information Services

C DE F G

Please send me your publication entitled "Honeywell Data Communication Capabilities."

A full complement of software is provided to handle
communications for any type of application. This software includes those routines for interrupt handling,
real-time input analysis, output stacking, random access
storage, line utilization and determination of line availability, and data protection. All communication software
may perform in conjunction with Series 200 Operating
Systems.
MAKE YOUR OWN COMPARISON
The following table lists important characteristics in a
data communication system. Honeywell's capabilities
can be compared with those of any other system by
filling in the appropriate data in the blank column provided.
Check List of
Data Communication Features

Honeywell

CENTRAL PROCESSOR
Simultaneous production and
communications?
I/O interrupt?
Memory protection features?
Program compatibility for backup?
Small-unit modularity?

Yes
Yes
Yes
Yes
Yes

INTERFACE
Single- and mUltiline interfaces?
No. of lines per multiline interface?
Character and message modes?
Maximum line speeds: Single-line
Multiline
Gradually expandable line-handling
capability?
Automatic switching for backup?

Yes
Up to 63
Yes
5100 cps*
300/7,000 cps
Yes
Yes

SOFTWARE
Communication-handling routines for:
Interrupt?
Real-time input analysis?
Output stacking and interfacing?
Random access storage and retrieval?
Line status?
Data protection?
Choice of operating system control?

Yes
Yes
Yes
Yes
Yes
Yes
Yes

*Higher speeds on special order

WRITE FOR MORE ON HONEYWELL
COMMUNICATION DATA CAPABILITIES
Honeywell's concept of dimensional data communication insures an exceptional opportunity to control costs
by fashioning your system to the dimensions of your
workload. For more detailed information on Honeywell
communication capabilities, complete and mail the accompanying coupon.

Name _________________________________________
Title_________________________________________
Company_______________________________________
Address
Clty ______________ State _ _ _ _ _ Zip Code _ __

Other

Honeywell
ELECTRONIC DATA PROCESSING

c&a
CAPITAL
REPORT
A Special Report from C&A's
Washington Correspondent

The management of computers in the Federal Governmcnt has taken a step up with the creation of a new ADP
(automatic data processing) Management Branch in the
Bureau of the Budget. Named to head the new Branch is
Joseph F. Cunningham, former associate director of data
automation at Air Force Headquarters in the Pentagon and
a sl~as()ned hand in Government computer circles.
Tlw new ADP Management Branch, one of the first concrete results of passage of the Brooks Bill, has four specific
dutil'S:
1.. To provide direction and guidance to Government
agencies in their management of data processing, and
to evaluate their efforts.
~. To construct a Government-wide information system
on data processing equipment. This will be a big
cxpansion of the old annual computer inventory.
:{. To push standardization of data elements and codes
among Federal agencies. Here Cunningham's extensive work with CODASYL, the Conferon Data Systems Languages, should be a definite asset.
.1. To promote the advanced design of data processing
systems, including the integration of systems on an
intra-agency and inter-agency basis.
I Jl addition· to Cunningham, the staff of the ADP Branch
includes J. P. Kingston, Clark R. Renninger, Fred J. Svec,
and John L. Little. Its nucleus came from the former group
of computer experts that functioned out of the limelight in
tlw Budget Bureau for so many years. This group worked
\lIukr William Gill, who· retired from Government service
at tlw end of 1965 and is now a consultant in Washington.
Slate governments generally lag behind the Federal Government in adopting some form of central management of

data processing equipment and operation, according to a
report issued jointly by the Council of State Governments
alld the Public Administration Service.
Of the 31 states taking part in a survey of computer
o\H'rations, only 11 said they had legislative or executive
din~('tives dealing with the organization or role of a central
data processing service. This is in contrast to the Federal
(;O\'(~rnment's 1965 computer legislation, which gave the
National Bureau of Standards responsibility for establishing
(,Olllplltcr centers and computer sharing, and' with the many
Budget Bureau directives issued since computers first entered
(;overnment service.
COMPUTERS and AUTOMATION for April, 1966

California leads all other states in the number of computers in use. Out of a total of 216 computers in all reporting states, it had 30. This figure does not include those
on state university and college campuses. California has
found 797 separate applications for its 30 computers. Looking at all states) the most popular areas for computer use
are education, highway engineering, and administration.
Copies of the report can be obtained for $5.00 each from
the Public Administration Service, 1313 East 60th St., Chicago, Ill,
Defense Secretary McNamara's "planning-programmingbudgeting" system is being adopted by civilian agencies, and
one of the long-range results should be the automation of
some part of the budget process. A few years ago a study
group concluded that the Library of Congress could not be
automated completely, and this may be true of the budget;
but even if it were proved conclusively that all of the budget
could be automated) barriers would be raised in Washington
to such an attempt, because the yearly review of the budget
in appropriation hearings is the basic form of communication
between Congress and the Executive agencies, It is not likely
that Congress would want these conversations to cease.
On the other hand, the automation of an agency's budget
before it is presented to Congress seems to be finding favor.
In February, the Pentagon announced that Bunker-Ramo Corporation is automating part of the Army's $11 billion budget.
Lt. Col. Richard Wormer, chief of the Army's Data Processing
Systems Branch, said this automation will be arrived at by
assigning an order of importance to the Army's various missions .. Then funds will be allocated to the projects according
to this order.
"The end result," Wormer said, "will be a more efficient,
effective, responsive, economical system designed to free the
various managers of details so that they can better exercise
their imagination and creativity."

JAMES TITUS

USING A COMPUTER TO
DESIGN A COMPUTER

Kathe Jacoby and Armand R. Laliberte
Philco Corp.
Subsidiary of Ford Motor Co.
Willow Grove, Pa.

As the computer industry has become more competitIve,
design and production methods which reduce the lead time
from functional specification to deliverable unit become increasingly important. Industry is more and more turning to
the computers themselves as useful tools for automating the
design of new computers. It is the purpose of this article
to explain and give evidence suppor'ting the use of computers to design computers, which we shall call "design
automation."
Design Automation
Design automation is the use of computers in the design
and production of new computers. When a computer is
designed, the process is usually as follows: a functional specification, a type of circuitry, and a type of pac~aging are
decided upon. The registers and their interconnecting gates
are generally specifie'd by a block diagram. The instructions
may be described by flow charts describing the action of the
elements of the block diagram. Alternatively, the instructions
may be described by equations. While the block diagram is
usually generated first, it is modified by feedback during the
definition of the instructions. Once this system logic is complete, the detailed logic is defined always acccording to the
rules required by the circuits and, in many cases, by the
packaging. The form this design takes is a matter of preference but is most likely to be in the form of either equations
or logical diagrams. It may be in the form of relationships
between building blocks that are considerably larger than
simple Boolean functions. Whatever the logic building blocks
may be, they must next be translated into physical building
blocks. This assignmeut is referred to as "placement". Design automation programs may perform some or all of the
placement.

Once the logic building blocks have been assigned to
physical blocks, the nodes (pins) that should be electrically
common can be connected. Most current packaging techniques effect these connections via backboard wiring. A set
of direct wire segments between nodes is selected by algorithms that depend upon circuit and packaging constraints
and that try to achieve optimization according to criteria
similar or identical to those for placement. If point-to-point
wiring is used, the wiring net is completely ·defined. Otherwise, either a fixed path is computed for each segment, or a
suitable path is routed if buildup of wire or crosstdlk are
problems. The documents required by production and test
departments can be generated at this time. Also if production
techniques include automatic or semi-automatic machines for
wiring and / or testing*, the design automation system should
include the programs for generating the appropriate input
media for this equipment. After a design is complete, it is
likely to be changed due either to design errors or to new
features added to a design. At this time, all documents must
be updated. Computers have proved themselves to b~ economically justified as an aid to many of these tasks.
Consistency
Design automation can enforce consistency in several ways.
If the design programs derive placement and wiring from the
logic, no inconsistencies can arise. Any inconsistencies among
logic, placement, and wiring generated by manual methods
will be detected by the checking of a design automation
system.
* Examples of automatic tools at the present time are the
Gardner Denver automatic wiring machine, continuity
testers, and semi-automatic wiring machines such as the
Philco Wiring Verifier.

COMPUTERS and AUTOMATION for April, 1966

Error Checking
Design automation permits error checking of various types.
The main error checks are that no logic, circuit, or wiring
rules be violated. While some transcription and keypunch
errors can be found by the programs, there must be some
checking by the designer that the logic, described by the input
and rcturned to him in convenient forms, is the logic he
intended. Simulation of the logic may be used to verify that
the logic yields the expected results.
Design automation greatly reduces the lead time between
the start of a design and the delivery of a computer product.
The wiring of a unit is generally in progress within a day or
two of completion of logic. Let us assume that keypunching
requircs a half day for a unit of about 700 logic elementl;,
and that turnaround time at a computer service center is
about four hours. An initial run to reveal errors requires
about twenty minutes of computer time. The evaluation of a
checking run may require from a couple of hours to a day,
depending on the thoroughness with which the designer compares the computer output to his initial design. Then thirty
to forty minutes are needed to run a complete initial series.

Why Design Automation?
Why usc design automation? The most convincing answer
to management is "it saves money". The advantage of lead
timc in getting a new product on the market quickly is
difficult to estimate in either dollars or percentage of total
cost. A contract may even require very short delivery time as
a condition for award. There are three areas where time is
saved: design, production, and debugging. The cost of the
design time is reduced because design is supported' for a
shorter time. Because production can use automated or semiautomated techniques, the time and cost for production is
reduced. Because the production documents are ~ccurate
and the production and debugging documents agree with the
initial design, debugging proceeds more quickly and less
expcnsively.
Even more significant than help in the development of the
prototype are the savings when changes are made. Changes
are implemented only after checking, and a complete and
continuous history of the design may be retained.

Example of a Design Automation System
. Typically, design automation systems take, as input, logic
m terms of relationships of building blocks, as shown in
Figure 1.
The representation of the relationships is a matter of con~eniencc; design automation systems, therefore, accept as
mput whatever form is usual for the desi ners involved.

Fiqure 1 - Each of ..l! to i is a building block, which might
be an inverter, an and-gate, or some other circuit unit.
The relationships may be described: (U in terms of input,
for example, i has inputs ~ and~; (2) in terms of driving,
for example, ~ drives ~ and i; (3) in terms of logical relationships, for example, if c were an inverter and f a~
and-gate, then i equals ~ and-not-..l!'
-

COMPUTERS and AUTOMATION for April, 1966

The building blocks can be represented by numbers or
names, or both. Again this consideration depends on the
convenience of the particular design group.
After the information that describes the relationships among
the logical building blocks has been entered into a design
automation system, information is checked by the computer
programs for any violations of logic or basic circuit rules.

Output
The information can be returned to the designer in different
forms. One form may be information describing both inputs
to and loads from a logic element. Another form may be a
set of input equations. Still another form may be equivalent
to flow charts of instructions. There are other forms of logic
presentation that are equally -convenient.
If the correctness of the logic is to be checked by simulation,
the simulation programs may be introduced at this point.

Criteria
The criteria for acceptability of a placement (i.e., correspondence between logical building blocks and location on a
physical building block) are varied.
Typical criteria are:
1. Minimize total wire.
2. Maximize the wiring i!1 a given physical direction.
3. Minimize the spread of source-to-Ioad distances.
4. Orient special circuits near the external interface.
5. Have registers aligned.
6. See that no wiring rule is violated. Wiring rules
depend on the circuit limitations specified by the
circuit designer. Typical wiring rules are as follows:
No wiring run may be longer than X inches. There
may be no more than N wires of type Y (plain wire,
twisted pair, for example) brai-Iching from a node as
a function of the circuit at that node. If there are W
loads on a wire run, the run may be no longer than
X inches.
7. Minimize the number of twisted pair wires.
8. Minimize the longest wires.
In most applications more than one criterion applies.
Some of the criteria involve reliability, some involve speed,
some involve cost. For example, if a wiring rule is violated
in the fabrication of a unit, this unit may not be adequately
reliable; hence this criterion is always important. Speed may
be increased when the spread of source-to-Ioad distance is
kept minimal. The spread of source-to-Ioad distance is the
difference between the maximum and the minimum sourceto-load distance. The spread of the lengths of wire runs
may be equally important. If delays through wire can be kept
standard, then total delay per stage can be held to closer
tolerances. The timing tolerances, which are considered in
both worst case and statistical design, are th.en reduced.
Cost is a reason for wanting to reduce the number of
twisted pairs. Reliability is the consideration, for example,
when special circuits involving connections that may generate noise, must be kept away from other wires. Cost is a
factor when special production techniques favor the maximization of the number of wires in a particular physical
direction. Cost is also a factor when maintenance and debugging operations are assisted by aligning registers when
indicators are present on the physical package.
It is generally difficult to apply all the desired criteria directly and it is especially difficult for the human designer to
maintain absolute "fairness" in multiple trade-off situations.

Computation

Arrangements of List
Initial List

Figure 2

(iI l!

.£.

lliJ

2- .£.

J! e 1.

[£)

a l!

.£.

a E. .£.

(2 + 3)/2 = 5/2 ~ 2

(2 + 3 + 4 + 6) /4 = 15/4 = 3

+ 3 + 5 + 6)/4 = 15/4 = 3

@J e 1.

(2 + 6)/2 = 8/2 = 4

~ 1

(3 + 6) /2 = 9/2 = 4

I1J

(2 + 3 + 4 + 5)/4 = 14/4 = 3

,
I

List after
1st iteration

After the logic has been checked, any placement information that the designer wants to specify may be introduced
and verified for consistency and the remaining logic (which
may be the whole or any portion of the design) can be
left for placement by the program. For example, many of
the placement programs used at Philco contain two parts.
The first part depends exclusively on the logic building blocks
and is independent of the nature of the building blocks or
the type and organization of packaging techniques. The
second part assumes that the packaging building blocks contain a number of logic building blocks that are generally not
interconnected on the package. The specific program depends
on the specific standard packaging techniques used.
The essence of the placement method is the reduction of
the logic elements to a linear list. This list is then manipulated to form linear clusters of associated elements. The first
phase examines the relationship of all of the logical elements
and then lists each element as dose as possible to its associated clement (s) .

SIGNAL
IDENTITY

INPUTS
GATE CONNECT

UNIT ...• TST

OUTPUT
GATE CONNECT

Consider the logic sketch in the example shown earlier
in Figure 1. We can make a list of building blocks:
Building Block List

Placement

GATE WIRING REFERENCE

a
b
c
d
e

We can also make an association list. Two elements are
called associated if a wire path exists between them.
Association List
a
a

b
b
b

c
a
c
d

c e
c f
d b
d f

b f

DATE .• 6-13-62
COLUMN
TIE
GATE CONNECT

e c
e f
f b
f c
f d
f c

a
b

PAGE 003

REV •• B

LOADS
GATE CONNECT

FROM

TWIST

LENGTH

B20-14

B20*1O

E31-12

B20*10

E31-12

13 3/4

36 66006

B21-02

B21*07

E32-05

B21*07

E32-05

13 1/4

E32-02

147

031-02

B23-05

E32*04

031-02

E32*04

54 B23-05

34 • E32-o3
36

E32-05

E32-o7

E32-o6

4 1/4

Figure 3 -- Sample Of Typed Copy Of Machine Printout

COMPUTERS and AUTOMATION for April, 1966

The algorithm then considers each element on the logical
building block list in turn. The subject element in each line
of Figure 2 is indicated by enclosing it in a box.
Each subject element is considered in relation to all its
associated elements. All the associated elements in Figure 2
are' underlined. The sum of the positions in the list of the
associated elements is divided by the number of associated
elements. The quotient represents the center of gravity of the
associated elements. The subject element is moved to this
position and all the elements between its initial and final
position are slid over one place.

LOG IC -ORIENTED SECTION:

I
I

I
I
,--------,

.- - -,- - - -'!

ERRORS

Iterations

I
I
I

I
CIRCIJTT-ORIENTED SECTION:

I
I
I
I
I

L __

Figure 4 -

Flow Chart Initial System

When every element has been considered as a subject
element, one iteration has been performed. For example, for
a device with about 1500 logic elements, about eight or nine
iterations are suitable. This example illustrates techniques,
but is not sufficient to show the merit of this method.
The list of logic elements is then assigned to physical
modules in such a manner that the order is disturbed as little
as possible. The backboard positions are considered as one
dimensional by ordering to left in one row and to the right
in the next.
After placement of the logic, the design automation system
determines the wiring and generates the documents required for fabrication and test. It also supplies information
on the correct media for automatic or semiautomatic production or test equipment. Figure 3 shows a typical debugging
document. Figure 4 illustrates the flow of a typical system
for initial design. Figure 5 shows a typical system for changes
to an existing design.
When a change is to 1)1" introduced, there are a number
of basic investigations to be made. Is it reasonable? Is
something being added that already exists in the file? Is
the change consistent? Does the wiring change agree with

(Please turn to page 58)

Release Unit
Date and Document
Numbers
Essential Files
for
Initial Design
Revision A Unit
Date and Document
Numbers
r
I

Essential Files
for
Revision A

'--------.-1

I
I
I
I
I

L._

Engineering
Documents

Tape
Motion

Revision B Unit
Date and Document
Numbers
Essential Files
for
Revision B

History
Figure 5 -- Flow Chart Revisions and Revision Tape
COMPUTERS and AUTOMATION for April, 1966

A NEW' integrated-circuit,
core ·memory system
from ·FABRI-TEK

II _ ... ........,.;..

Here is a compact, versatile memory system at
a truly economical price "Yhich can perform:
any, of the standard data storage functions
with reliability. Full cycle time is 2 microseconds, half-cycle time is 1.25 microseconds.
Access time is 850 nanoseconds. Four access
modes are possible: Random; Sequential;
Random/Seguential; and Sequential-interlaced. Capacities available are: 64, 128,256,512,
1024, 2048, and 4096 words, with 2 to 30 bits per
word in increments of 2 bits. A choice of input
and output interface circuits and optional address register is offered. Power supply and selftest exerciser are available options. Fabri-Tek's
•'standardized design" concept gives you a

custom fit to your particular memory requirements with the economy of mass production.
This new memory system from Fabri-Tek can
be ordered ,out of a catalog. You choose the
features by the number and your system will
be delivered, ready to plug into your equipment
rack. A oonnector wiring diagram of your
specific system will be sent to you before delivery of the memory system so you'll be ready
to 'operate without waste of time.
Ask for Bulletin 653.5, Series MUA2 Integratedcircuit core memory system. Write, call, or
wire: Fabri-Tek, Incorporated. Amery, Wisconsin; phone 715-268-7155; TWX 510-376-1710.

FABRI-TEK LEADS IN MEMORY TECHNOLOGY
Check WIth Fabri· Tek for rewarding engineering opportunities!

Designate No. 2 on Readers Service Card

COMPUTERS and AUTOMATION for April, 1966

UACROSS THE EDITOR'S DESK"
Computing and Data Processing Newsletter

TABLE OF CONTENTS

Applications
New Contracts.
New Installations.
Organization News
Education News •.
New Products.
Memories.

Data Transmitters
and AID Converters •
Input-Output •
Components •
People of Note.
Meeting News •
Business News •

.41
.43
44
45
46
47
. 47

.48
49
50
. . 52
.52
.53

APPLICATIONS

MOLECULAR STRUCTURE
"PICTURED" BY COMPUTER

Scientists at Argonne National
Laboratory (operated by the University of Chicago under contract
to the Atomic Energy Commission)
have "taught" the Laboratory's sophisticated computers to display
in pictorial form amounts of data
so vast that they would be virtually unuseable in mathematical
form. This accomplishment permits
scientists who are studying molecular structure to compare various
models of two-atom molecules' by
looking at "pictures" of their
structure instead of trying to
compare large numbers of complex
tables.
Dr. Arnold C. Wahl, now an
Assistant Professor in the Department of Chemistry and in the Theoretical Chemistry Institute at the
University of Wisconsin, but formerly of the Argonne Chemistry Division staff, prepared computer
programs which instructed a computer to construct and "draw" contour diagrams of molecular orbital
and total electron densities for a
number of molecules. He was assisted by members of Argonne's Applied Mathematics Division staff.
Physicists and chemists have
been attempting to construct an
adequate mathematical model of the
structure of molecules for many
years. However, the complex differential equations required to define such a mathematical model are
extremely difficult to solve. In
1928, a significant advance occurred when R. S. Mulliken and
COMPUTERS and AUTOMATION for April, 1966

F. Hund developed for molecules
the orbital theory, or "shell
model". Although this orbital concept simplified the problem greatly,
calculating the orbitals still required a prohibitively large amount
of algebra and arithmetic.
In recent years, the availability of high speed computers increased the usefulness of this orbital theory because it became
practical to carry out this large
amount of algebra and arithmetic
and to calculate the exact properties of the Mulliken-Hund model
for molecules consisting of two
atoms. As a result of these calculations, the model is now well
documented and its usefulness is
being demonstrated.
Dr. Wahl points out, however,
that if the advances and refinements of this model, which have
been made possible by high-speed
computers, are only to be descri bed
in complex mathematical language
or in terms of vast, undigestible
and often misleading numerical
tables, they will have limited
value.
In order to eliminate this
obstacle to widespread use of the
molecular model both in teaching
and research, Dr. Wahl set out to
prepare programs which would enable a computer to present a visual representation of the model -a picture which is correct, and is
much more complete than anything
which has thus far been available.
The completely automatic programs
developed by Dr. Wahl have made it
easy to analyze visually a very
large amount of calculated data.

They are already being used in new
studies of molecular structure. In
one such investigation, a study of
interatomic forces and the formation of the chemical bond, these
computer programs ~re being used
to display changes occurring in
electronic charge density as a
molecule forms.
Although the initial application of this new tool has been to
studies of molecular structure, Dr.
Wahl expects it to be useful in
many areas where large-scale computational efforts result in such
vast amounts of data that only a
small fraction of it can be analyzed by conventional methods.
Programming a eomputer to enable
it to display such data visually
and thus communicate directly with
one of man's most sensitive and
subtle senses can greatly increase
the effectiveness of the interplay
between man and computer, Dr. Wahl
said.
Dr. Wahl's work was reported
in the February 25th issue of Science Magazine.

MARKET DATA SYSTEM USES
OPTICAL MARK READING

A card dropped into a compact
gray box on the floor of the New
York Stock Exchange has set in motion a major technological breakthrough -- the use of optical mark
reading to capture continually
changing information. The electronic first was scored as the Exchange prepared to place on-line
its fully-automated Market Data
41

Newsletter
System. The system was specially
developed by IBM Corporation as a
major part of the Big Board's
broad-scale automation program.
In the initial phase of the
program, the specially-designed
cards are being used to capture
trading data at Post 17 on the Exchange floor, where 119 of the more
than 1600 Big Board listed stocks
are traded. As a trade takes place,
the reporter, using an ordinary
lead pencil, records the details
by drawing lines through speciallycoded boxes on the pre-printed card
designating the stock symbol, number of shares, and price.
REPORTERS CARD
Stock Identification

r--::::::OJ ~ [iii] [iii] I£!!J ~ ~
~ m@i]~I!!!lI!iil[MQC]§]@!}

IT)

OlUME ~j~~~;;~~'

Second
Trade

011

5678

COMPUTERIZED
BOOK CATALOG

Stanford Uni versi ty' s $5 mi Ilion
undergraduate library (Stanford,
Calif.), scheduled for completion
next fall, will have a computerized book catalog, Stanford Librarian Rutherford D. Rogers noted in
his annual report. "This is the
first major step toward computer
application to library technology
at Stanford," his report said.
"Stations" for instantaneous
consultation of the list of books
available in the library will be
located in each major room of the
new library and can be extended to
the departmental offices of the
University, Dr. Rogers said. This
system will allow much quicker
scanning of available holdings and
will allow for quick copying of
the list of books on a particular
subject for distribution to students, faculty members and research scholars.

Third

Trade

Bid-Asked
Quotation

Card
Identification
Code

The card, which provides space
for three trades and a bid-asked
quotation, then is placed in the optical card reader located at the
post. The optical reader scans
the pencil marks and transmits the
information in a fraction of a second to the computer center three
stories above the trading floor.
Here, the data is checked automatically for validity, processed, and
recorded magnetically to provide
information for the Exchange's
Telephone Quotation Service. Within seconds, a computer-assembled
announcement of prices is made via
loudspeakers mounted at the post,
and the transaction is reported
at the post on a special printer.
As card readers are being installed at the 19 trading posts on
the floor, the present manual reporting and ticker operation for
each of the posts will be phased
out. When fully operational later
this year, the system will speed
trading information out to the Exchange's international ticker network, usually within a fraction of
a second after it has been reported
on the floor.
42

DIGITAL COMPUTER CONTROL
APPLIED TO
CEMENT MANUFACTURING

Cement manufacturing entered
a new era with the recent announcement by Dundee Cement Company (Dundee, Mich.) that it has successfully applied a digital computer
to the control of a giant rotary
kiln. Roblee B. Martin, Dundee
president, said the firm's 460 x
16 foot rotary kiln at its Dundee
plant is now being controlled on
a "closed-loop", or fully automatic, basis by an IBM 1710 process
control system. This achievement
culminated more than a year of
research and development by Dundee
and IBM Corporation.
"Our experience wi th closedloop kiln control has clearly demonstrated the ~ystem's ability to
maintain temperatures at pre-set
levels throughout the 460-foot
length of the kiln," said Werner
Ostberg, Dundee vice presidentoperations. "Temperature control
is the most critical aspect of the
cement manufacturing process," he
said. Significance of the achievement was explained by Mr. Ostberg,
in terms of potential production
gains, reduced maintenance costs
and uniformity of product.
The computer, which can make
thousands of calculations a second,
instantaneously reacts to fractional degree temperature changes.
To control the kiln operation, the
computer gathers information --

temperatures, speeds, pressure,
etc. -- from about 70 instruments
located along the kiln. Based on
predetermined standards of quality
control, the computer calculates
what simultaneous adjustments, if
any, must be made. The computer
then automatically positions the
kiln actuators -- valves, dampers,
fans, etc. -- to new settings which
will maintain the process variables
at pre-set levels.
Setting this system apart from
other process control systems is
its ability to consider the state
of the entire kiln before making
any adjustments.

BRITISH TOWN USING
COMPUTER IN BRIDGE DESIGN

A computer is being used to
help design a ~ridge over the River
Aire. The bridge, to be constructed by cantilevering out from each
bank, with the two sections finally joined by a pin, is the first
of its type to be built in England,
according to Col. S. M. Lovell,
Yorkshire County engineer and surveyor. If conventional bridge design techniques were used, Lovell
said, they would destroy the
river's small navigable channel.
A Honeywell 400 computer installed at the county's headquarters several months ago is computing
the profile of each of the 64 sections of the bridge to correct
shrinkage, stress and creep problems that could prevent the bridge
from meeting exactly at midpoint
over the river. As many as 64,000
separate calculations are required
to correct just one part of a section's profile, Lovell said. In
addition, thousands of other calculations are needed to estimate
stress and deflection in preformed concrete beams, determine
distribution of loads between
beams, and aid the design group in
selecting quantities and shapes of
steel and concrete reinforcements
needed in the bridge.
Col. Lovell said the county
plans to use the computer in constructing roads, by having it provide data on proper road curvature
and camber. It also will aid in
producing perspective drawings that
permit road and bridge designs to
be viewed from different angles -a technique used by engineers to
correct misalignments and other
design errors.

COMPUTERS and AUTOMATION for April, 19GG

Newsletter
MESSAGES RELAYED TO AND FROM
TIME-SHARING COMPUTER
BY SATELLITE

A General Electric computer
has received and answered messages
relayed via satellite over a
200,OOO-mile round-trip route that
touched on two continents. Simultaneously it handled unrelated
problems fed into its multi-tracked
"brain" from nearly 40 other scattered locations in the United
States. The demonstration, recently held in Cleveland, Ohio, marked
the first time messages have been
relayed to and from a time-sharing
computer by satellite.
The Early Bird satellite, stationed 22,300 miles above the midAtlantic, was used as the link be-

tween the United States and France.
The Communications Satellite Corporation, in cooperation with its
counterpart in France, made available the satellite circuits. Impulses traveled at the speed of
light -- 186,000 miles a second.
Two time-sharing computers
were used to receive and reply to
the messages from Cleveland -- a
GE-225 in Oklahoma City and a
GE-265 in Phoenix, Ariz. Signals
traveled to the two computers by
telephone lines from Cleveland to
the General Electric Computer Center in New York, then to ITT World
Communications headquarters in New
York, then to the Comsat-operated
earth station at Andover, Maine.
From the transmission center in
Andover they were flashed by mi-

crowave to Early Bird, then to a
receiving station in France, then
back to Early Bird and back to Andover. From Andover the signals
traveled by wir.e again to ITT in
New York and then to General Electric's computer headquarters in
Oklahoma City or Phoenix.
"This dramatic accomplishment,"
said Dr. Louis T. Rader, vice-president and general manager of G-E's
Information Systems Division, "illustrates again the virtually unlimited potential of time-sharing
computers." He predicted that, ultimately, time-sharing computers
will be capable of becoming worldwide storehouses of information
that can be tapped instantaneously
from any point on earth.

NEW CONTRACTS

Western Union

SCM Corporation, New
York, N.Y.

Puerto Rico Water Resources
Authority, San Juan, Puerto
Rico

The Foxboro Company, Foxboro,
Mass.

Alitalia Airlines, Rome,
Italy
Department of Defense, Air
Force Rome Air Development
Center, Griffiss Air Force,
N.Y.
Sauena Belgian World Airlines. Brussels. Belgium
United States Air Force,
Rome Air Development Center

General Precision, Inc., Link
Group
University of Illinois at
Urbana

Aeronautical Systems Division,
Air Force Systems Command

Sylvania Electric Products,
Inc., a GT&E subsidiary

General Services Administration
Santa Barbara County, Calif.

LMC Data, Inc., New York

General Precision, Inc., Link
Group
General Electric Company

Cubic Corporation, San Diego,
Calif.

Lower Colorado River Authority, Austin, Texas

General Electric Company,
Instrument Dept., West
Lynn, Mass.

Prince George's County,
Upper Marlboro, Md.

Radio Corporation of America

Jet Propulsion Laboratories,
Pasadena, Calif.

Scientific Data Systems,
Santa Monica, Calif.

U. S. Navy, Bureau of Yards
and Docks, Washington, D.C.

Honeywell EDP, Wellesley
Hi 11 s, Ma s s •

COMPUTERS and AUTOMATION for April, 1966

Electronic data communication sets for
a new high-speed communications terminal
for industry and government
Instrumentation for Units 9 and 10 of San
Juan Steam Plant. The electronic control
systems for Units 7 and 8 of the same station also were supplied by Foxboro
DC-9 digital flight simulator
Designing and building world's fastest
and most advanced computer. Device will
be known as Illiac IV and is planned for
upwards of 1 bi Ilion computations a second
A Link 727 digital flight simulator

$2 million
over $1 million

over $1 million
$8 million

over $1 million

A large-scale GE-645 computer to help
in its advanced electronics development work in computer technology -delivery scheduled for 1967
$3.2 million
Manufacture of components and spare
parts for advanced airborne data
processing eguipment
Leasing data processing equipment and
for service
Delivery of 18 Votronics vote counters
over $490,000
plus collateral equipment -- "lease with
option to purchase" contract extends for
five years
A second GE/MAC (General Electric/Measure$187,000
ment and Control) system to be used in connection with a 125-mw turbine-generator set
at the Authority's Sim Gideon plant, Bastrop,
Texas
Provision of a computer facility to
eventually consolidate all County record
keeping in one system -- target time for
operation of central computer system is
mid-summer. 1967
13 SDS 920 computers to be used by the
$1,867,000
NASA-JPL deep space network in forthcoming space exploration projects
Lease of one H-200 computer system at the
Public Works Center, San Diego, Calif., to
be followed by similar installations at 8
other Public Works Centers around the world
-- equipment valued at about $1.5 million

Newsletter

Western Geophysical, Los
Angeles. Calif.
Emery Air Freight, Wilton,
Conn.

Redcor Corporation, Canoga
Park. Calif.
IBM Corporation

Headquarters, Department of
the Army

The Bunker-Ramo Corporation,
Sil ver Spring, Md.

Iranian Oil Refining Company,
Abadan, Iran

Bonner & Moore Associates,
Inc., Houston, Texas

F·ield digital seismic recording systems

over $2 mi Ilion

Real time computertized air cargo data
ne'twork -- system will include two IBM
System/360 Model 40 computers
Automating a portion of their budget
system -- automation of Army budgetary
processes is being done to institute
economy into Army's budget by allocating funds to projects in their
order of importance
Development of an automated pipeline
scheduling program which will facilitate scheduling of movements of products from the refinery to the Bandar
Mashur terminal on the Persian Gulf
and of the operation of the terminals

$3 mi Ilion
$348,000

•

NEW INSTALLATIONS

OF
R. J. Reynolds Tobacco, Co.,
Winston-Salem, N.C.

IBM System/360 Model 30 valued
at $400,000

Heights Academic Computing
Facility (HACF) , New York
University. New York. N.Y.
Curtis Publishing Co., Philadelphia, Pa,

IBM System/360 Model 30

Gimbels-Pittsburgh, Pittsburgh, Pa,
McDonnell Aircraft Corp.,
General Engineering Div.,
St. Loui s. Mo.
Intinco Ltd., London,
England
C-E-I-R, Inc., Washington
and New York service bureaus

Two Control Data 8090 computer
systems with Control Data 915
page readers
H-200 system

Model SCS 670-2 computer
UNIVAC 418 computer
Three IBM System/360 Model 30
computers having a total value
of more than $1 million

Sladtische Sparkasse Bochum,
Bochum, Germany

NCR 315 system; a second
system is on order

Space Flight Operations
Facility, Pasadena, Calif.

Four S-C 3070 high speed
pri nters

Jersey-Kapwood Ltd., (manufacturer of fabrics, lace,
lingerie), Nottingham,
England
Bankers Data Corp., Chicago,
Ill.

Honeywell 120 system

GE-415 system

Dutch Statemines, Limburg,
Netherlands

CAE 510 computer system

Bowring insurance group,
London. England
Duncan Electric Co., Lafayette,
Ind.

Honeywell 400 computer system
Control Data· 160 computer
system

IBM Corporation
Increasing company's capabilities for
computer processing; previously installed IBM 1401 and new equipment process accounting, keep track of warehouse inventories nation-wide, handle
stockholder records, and calculate for
industrial engineering studies
IBM Corporation
Student use by School of Engineering
and Science apd the University College of Arts and Science
Control Data Corp.
Processing firm's huge volume of
subscriptions
Data processing activities involving
all areas of merchandise analysis;
other applications will include business accounting and sales analysis
Inclusion in high speed digital data
acquisition system

Honeywell EDP

Scientific Control
Systems, Inc.

Operating hub of SCAN (Stockmarket
Sperry Hand Corp.
Computer Answering Network)
IBM Corporation
Association, financial, business and
government clients at Washington;
two machines were added to firm's
equipment complex in Manhattan
Bank's use initially in the automatic
National Cash Regprocessing of customer "standing
ister Co., Internaorders" (refers to customers' instruc- tional Division
tions to deduct certain amounts each
month from a savings account and apply
funds elsewhere -- common in Europe)
Stromberg-Carlson
Operation from IBM 7044 and 7099 computers; will print information conCorp., Data Products
cerning JPL's unmanned planetary,
Div.
interplanetary and lunar probes for
NASA
Initial use for cost estimating of
Honeywell EDP
pricing structures for firm's entire
product line -- other applications to
be added later
Demand deposit accounting, savings,
General Electric Co.
installment loan and other bank and
commercial accounting operations
Control of new 100,000 tons per year
Compagnie europ~enne
ethylene plant
d'Automatisme Elect ronique
Honeywell EDP
Hanuling accounting and management
statistics
Calibrating watt-hour meters to new
Control Data Corp.
levels of accuracy and reliability
COMPUTERS and AUTOMATION for April, 1966

Newsletter

DC Trucking Company, Inc.,
Denver, Colo.

GE-415 computer system

U. S. Navy, Pacific Missile
Range, Point Mugu, Calif.

Control Data 3100 system

American Cyanamid Co., Stam,ford, Conn.

SDS 925 compu ter

Mail Advertising Corporation
of America, Lincoln, Nebr.

IBM System/360 Model 30

The Outlet Company, Providence. R.I.
Interstate Power'Co., Dubuque, Iowa

NCR 315 system
Control Data' 636 Industrial
Computer System

NASA's Marshall Space Flight
Center, Huntsville, Ala.

IBM 1130 Computing System

Crocker-Citizens National
Bank, San Francisco, Calif.

Control Data 915 Optical
Page Reader System

Tridea Corp., a Subsidiary
of Conductron Corp., South

Two DDP-116 computer systems

Pasndt!na. Calif.

ORGANIZATION NEWS

HONEYWELL EDP TO MARKET
BUNKER-RAMO DISPLAY DEVICES

Honeywell's electronic data
processing division, Wellesley
Hills, Mass., will market The Bunker-Ramo Corporation's new Series
200 and 400 information display
and inquiry devices in commercial
and government data processing
markets here and abroad, it was
announced jointly by Walter W.
Finke, president of Honeywell EDP,
and John E. Parker, chairman of
Bunker-Ramo. Four basic models of
the high-speed cathode ray tube
displays will be offered by Honeywell in its product line, under
this arrangement.
Bunker-Ramo will continue to
manufacture and market these input/
output display devices. The worldwide service organizations of both
firms will coordinate field servicing of the devices.
The arrangement also provides
Honeywell with rights for future
manufacture of display units in
the United Kingdom, Germany and
Canada as well as in other selected
COMPUTERS and AUTOMATION for April, 1966

All business and management operations; will become hub of a computercontrolled communications network that
wi.ll tie together all of company's operations in this country and Europe
Programming and operation by personnel
of Geophysics Division of the Range
Operations Dept. in areas of meteorology. oceanography, and climatology
Aid in research leading to new chemical products including textile fibers,
plastics, refinery catalysts, and fuel
cell electrodes
Increasing firm's data processing
capabilities and opening new marketing areas in the direct mail advertising field
Computerization of accounting procedures
Monitoring and data logging at the
M. L. Kapp Station, Unit #2, Clinton,
Iowa -- as part of the SAL (Scanning,
Alarming, Logging) system, the 636
will monitor the status of all major
equipment in the power plant
Use in support of NASA's Saturn
launch vehicle lunar program; a
major application is assisting
in Saturn antenna design
Translating bank and customer raw
information from original source
documents into useable 'computer
language'
On-line machine' tool programming
in conjunction with a Trideadeveloped automatic line tracer

countries. The agreement excludes
Japan, where Bunker-Ramo display
devices are marketed by the Nippon
Electric Company. Nippon Electric
also manufactures and markets
Series 200 computer systems under
a Honeywell license.

ISC ACQUIRES ASSETS OF
CONSTRUCTION CONTROL, INC.

Information Sys tems Co., Los
Angeles, Calif., has acquired the
assets of Construction Control,
Inc., Anaheim, Calif., supplier of
management consulting services to
the construction industry, according to M. O. Kappler, President of
ISC. Terms of the transaction
were not disclosed.
SIMULATORS, INC. ENTERS
MARKETING AGREEMENT WITH
TELEFUNKEN A.G., GERMANY

Simulators, Inc., Northbrook,
Ill., has entered into an exclusive marketing agreement with Telefunken A.G., Ulm, Germany. Mr.
Charles J. McVey, president of
SimUlators, Inc. announced that
the company will have complete
United States marketing respons-

General Electric Co.

Control Data Corp.

Scientific Data
Systems
IBM Corporation

National Cash Register Co.
Control Data Corp.

IBM Corporation

Control Data Corp.

Computer Control
Co., Inc.

ibility for the Telefunken line of
analog and hybrid computing equipment. Telefunken was represented
by Dr. Lux and Dr. Zur.
Simulators, Inc. plans to use
the Telefunken computing e'quipment
with electronic mode control designed by the Automation Center of
McDonnell Aircraft Corp. and produced by Simulators, Inc. under
licensing agreement. A line of
scientific computers is to be introduced at the Spring Joint Computer Conference this month. By
joining the efforts of Telefunken,
the McDonnell Automation Center
and an internal development program
Simulators, Inc. wi 11 enter the scientific computer market having the
world's widest line of 10 volt reference analog computing equipment
and have initially a system of
field proven performance.

CSC COMPLETES ACQUISITION
OF GEONAUTICS, INC.

Computer Sciences Corporation
has completed the acquisition of
Geonautics, Inc., of Washington,
D.C. Fletcher Jones, president of
CSC said holders of more than 97
per cent of the 116,555 Geonautics
shares outstanding have tendered

Newsletter
their stock to CSC. The offer was
contingent upon the acceptance by
holders of 90 per cent of the Geonautics shares outstanding. The
transaction involved 9,017 shares
of Computer Sciences' capital stock.
Geonautics provides professional and technical services in
the fields of geodesy, oceanograph~
navigational systems and astronomy.
The acquisition of Geonautics will
further expand the capabilities of
Computer Sciences and its subsidiaires in the field of the information sciences. Geonautics will be
operated as a part of System Sciences Corporation, a wholly-owned
subsidiary of CSC located in Falls
Church, Va.

ITT AND ABC DIRECTORS
APPROVE CONTRACT FOR
FIRMS' MERGER

The boards of directors of
International Telephone and Telegraph Corporation and American
Broadcasting Companies, Inc., approved a contract in February covering the merger of the two communications firms. Before the merger
becomes final, stockholders of both
companies must approve. In addition, completion of the transaction
is dependent on approval by the
Federal Communications Commission
and on a favorable tax ruling. It
is expected that meetings of stockholders of the two companies will
be held in the latter part of this
month.
The formal contract, approved
in February, stemmed from discussions between ITT and ABC for several months during 1965. One of
the conditions of the merger will
be the continued autonomous operation'and management of ABC by its
present management as a separate
subsidiary of ITT.

CONTROL DATA ANNOUNCES
$120 MILLION LOAN AGREEMENT

William C. Norris, President
of Control Data Corporation, has
,announced that negotiations have
been completed with ten major banks
for a revolving credit agreement in
the amount of $120 million. The
term of the agreement is two years.
Norris said that the funds will be
used for increased working capital
needs, and to finance the increasing leased equipment business which
is present in Control Data's backlog and incoming orders.
·'16

LEASING OF COMPUTER SYSTEMS
FOR EDUCATIONAL PURPOSES

Fabri-Tek, Inc., of Minneapolis has announced a leasing plan
for its BI-TRAN SIX Computer Educational System. Under this plan
the system can be leased for one,
two or three months at a nominal
cost. During this period if the
leasee decides to buy the system
any rental fees paid are applied
to the purchase price. Arrangements for longer leases are also
possible with purcha~e options.
This plan provides the means
whereby educational institutions
can actually use a computer system
in classes to determine the desirability or applicability of using
this type equipment in existing or
proposed curriculum. Monthly
leasing has been purposely establishedat a low rate to give all
schools an opportunity to investigate computer training through
actual use of the equipment.
(For more information, designate
#41 on the Readers Service Card.)

EDUCATION NEWS

STUDENTS LEARN MATH FROM
COMPUTER 3000 MILES AWAY

Prof. Anthony Oettinger at
Harvard opens his class by turning, not to a blackboard, but to
the signalling device. On a keyboard, he types out a problem, thus
programming a computer 3000 miles
away in Santa Barbara, Calif. The
computer, empolying a systems program worked out by Prof. Glenn
Culler of the University of California, solves a problem in calculus or statistics, step by step,
for the Harvard class. The mathematical process appears, almost
instanteously, on a display device.
Three universities now are
using the systems program developed by Professor Culler at Santa
Barbara; UCSB, UCLA and Harvard.
At Harvard, the computer is being
used in a program of experiments
in the process of teaching and
learning. Research aspects are
supported by the Advanced Research
Projects Agency of the Department
of Defense.
Professor Oettinger is using
the system experimentally for instruction in a seminar on "Technological Aids to Creative Thought,"

as well as in other classes. The
communication over the 3000 miles
is made by telephone and microwave.
Professor Oettinger foresees similar
displays in many classrooms, all
linked to the computing center.
"For the first time relatively
inexpensive display devices can be
installed in classrooms without regard to where a relatively more expensive computer is located. This
means literally hundreds of schools
and laboratories are now possible
participants in a system of this
kind." Essentially this method involves translating computer calculations into graphic terms which
dramatically illustrate a problem
and its successive steps to solution. "It's a case of a picture
being worth a thousand words. Besides that, there are some problems
you simply cannot describe in
words," he said.

COMPUTER-TEACHER SYSTEM FOR
FIRST-GRADERS NEXT FALL
IN EAST PALO ALTO

First actual use of a computerteacher system in an American elementary school will get under way
next September at Brentwood School
in East Palo Alto, Calif. At that
time, 150 first graders will receive instruction in reading and
mathematical skills a half hour per
day, with the computer keeping
track of their scores, feeding new
materials as the students' abilities
increase, and analyzing the data so
that teachers and school officials
can keep a day-to-day check on each
student.
The system is an outgrowth of
joint research and experiments over
the past five years by Prof. Patrick
Suppes, director of the Institute
for Mathematical Studies in the Social Sciences at Stanford University and Prof. Richard C. Atkinson
of the Psychology and Education Departments. A model installation
was first demonstrated last March
in laboratory form at the Stanford
Computation Center.
For the new Brentwood firstgrade program, a new model computer
system and terminal, developed during the past year by IBM, will be
installed. It permits the pupil to
communicate with the comput,er by
light pen which he can touch to a
television-type screen, by typewriter keyboard, or by audible response to a tape-recorded voice
from the computer. Problems in
mathematics or sentences for reading are flashed on a screen about
COMPUTERS and AUTOMATION for April, 1966

J
f
.l

t. . '.'.'. .';

,:l.~

Newsletter

8~ x 11 inches in size.
The student responds by touching the light
pen to the correct answer, typing
it out, or answering verbally.

If the answer is wrong, the
computer tells the student audibly
and resubmits the question. The
right answer brings up the next
question, printed on an eight-millimeter film "chip" projected on the
s~reen at the computer's command.
The computer is programmed to increase the difficulty of the material as the pupil learns and improves. It also can be specially
programmed for those with learning
difficulties.
The project is supported by a
$1 million grant from the U.S. Office of Education.

NEW PRODUCTS

Digital

DII AN CONTROLS ANNOUNCES
NEW COMPUTER TYPESETTING
SYSTEM

DI/AN Controls,
Mass., has announced
a new development in
total composing room
setting systems.

Inc., Boston,
the Model LC-3,
their line of
computer type-

Built around a special-purpose
computer, the new Computer Keyboard
Model LC-3 has many improved operating features, in addition to being more compactly styled with
fewer set-up controls. Hyphena-

~
,

tion, justification and error correction are all completed before
the tape is punched. Idiot tape
is eliminated. No rubouts are encountered because the operator
COMPUTERS and AUTOMATION for April, 1966

produces hard copy and can quickly
detect and rectify errors. Corrections are made simply by using
one keystroke to erase either one
letter, one word or an entire line
at a time. The correct copy is
then typed with none of these corrections ever appearing on the
tape. The result is an error-free
tape complete with linecasting instructions.
(For more information, designate
#42 on the Readers Service Card.)
IBM SPACE COMPUTER
"VOTES" ON SIGNALS

The most advanced space computer yet developed is slated to
navigate the first 625-ton NASA
Saturn IB launch vehicle. The 88pound, suitcase-size device features triple sets of circuits that
are tested by "voter" circuits to
screen out erroneous signals, and
duplicate memories that reset each
other in case of error. The computer, developed by the IBM Corporation, and a companion device
called the data adapter are located in the IBM-produced Instrument Unit (IU) stage, a three-foot
"control center" placed just below
the spacecraft and atop the propulsion stage.
The equipment is designed to
take flight data from instruments
aboard the IU and to calculate exact courses for the Saturns. To
provide the highest reliability
possible, certain modules in the
units are triplicated. Sets of
triplicated signals are sent
through "voting" circuits which
are designed to block a signal if
it does not agree with the other
two signals.
Another feature is a redundant memory system. Identical
flight data is put into two memories before each flight. These
memories constantly check themselves for incorrect data which
might be caused by stray electrical signals. If an error appears
in the memory being used by the
computer, it is sensed and, in a
spli~ second, operation is switched
over to the duplicate memory. After this switchover, the erroneous
memory corrects itself using data
from the second unit.

Memories

NANOMEMORY 650LARGE SCALE MEMORY SYSTEM
BY ELECTRONIC MEMORIES INC.

A new, large ~cale memory system operating at a cycle time of
only 650 nanoseconds will be demonstrated for the first .time at the
Spring Joint Computer Conference
(see Meeting News) by Electronic
Memories. The system, called the
NANOMEMORY 650, has a capacity of
16,384 words of up to 84 bits and
an access time of 300 nanoseconds.
It is the fastest large core memory system in production, according to William Richmond, Director
of Marketing for the firm.
The NANOMEMORY 650 has 2 ~-D
selection techniques, a 20-mil core,
and high speed silicon circuitry.
The combination provides greater
reliability, reduction of system
noise, elimination of read-write
time separation, and wider operating margins. Cost per bit is comparable with memories operating at
a cycle time of 1.25 microseconds.
The device will be demonstrated
in a standard 19-inch rack mounting
configuration with a capacity of
16,384 words and a word length of
28 bits. There are two other
standard configurations and custom
configurations also are available.
(For more information, designate
#44 on the Readers Service Card.)

AUXILIARY PROGRAM MEMORY
EXPANDS CAPACITY
OF MATHATRON

A new auxiliary program memory
storage device, which greatly expands the computational capabilities of the Mathatron electronic
calculator/computer, has been announced by Math~tronics, Inc. of
Waltham, Mass. Called the APS (for
~uxiliary frogram ~torage), new
unit in conjunction with a Mathatron 8-48 desk top calculator/
computer provides 48 to 88 individually addressable storage registers plus 480 steps of program
memory.
Data, formulas, and programs
are entered directly through a
simple calculator type keyboard
using ordinary algebraic language
including parentheses. Up to 18
pre-wired progiams, of 48 steps
each, may be incorporated into the
·17

NewsleHer
The equipment includes a transport mechanism, two electronics
chassis and one power supply chassis all housed in a self-supporting
metal cabinet 6.5' high by 25" wide
by 30" deep. All cooling is incorporated so that in normal computer
applications no special temperature
control is required. Standard 115
volt 60 cycle power is used. Many
opti~nal features are available.
(For more information, designate
#45 on the Readers Service card.)

Data TransmiHers
and AID Converters

device and entered automatically by
the program or with a touch of a
button. Decimal points are located
automatically and all answer,s are
printed out to 8 or 9 significant
digit accuracy with a 2 digit power
of 10 exponent. In addition to the
storage registers and program memory, 5 separate registers are used
for arithmetic manipulations. Number capacity is 100 columns.
Optional accessory devices include a paper tape punch and reader
with page printer, remote location
keyboard and direct entry punched
paper tape reader.
(For more information, designate
#43 on the Readers Service Card.)

MAGNETIC CARD MEMORY
PROVIDES
100 MILLION BIT CAPACITY

Model MCM-l Magnetic Card Memory, developed by Computer Acces~
sories Corp., Santa Barbara, CalIf.. ,
p~ovides an on-line random access
st~rage capacity of 100 million
bits with a maximum access time of
50 milliseconds. The device uses
long wearing flexible magnetic
cards contained in interchangeable
cartridges as the storage medium.
Each cartridge holds 64 cards and
can be loaded or unloaded on the
MCM-l in less than five seconds.
A single 32-track dual-gap
magnetic head provides read-afterwrite capability and the maximum
data rate is 10.4 megabits/second
when 32 tracks are processed in
parallel. There are 128 recorded
tracks on each magneti~ card divided into four groups of 32 tracks
each. Moving the magnetic head
from one group to another takes
less than 20 milliseconds.

TALLY SYSTEM 311 FOR
DATA TRANSMISSION

The new System 311 for data
transmission, by Tally Corporation
of Seattle, Wash., makes,high-speed,
error-free communication between a
central computer and outlying
points economical for the "data
transmission mass market"
At a
recent press conference in New
York City, Tally Corporation officials described the system as a
development which "will help industry realize the full potential
of computerization."
Each System 311 terminal incorporates a paper tape perforator,
reader and electronic logic. The
Tally System 311 Send/Receive Terminal transmits or receives data at
speeds of 1200 words per minut~
over ordinary dial telephone lInes,
will operate completely unattended,
detects and deletes errors, and
can be used for off-line tape duplication and editing in its spare
time.
According to company spokesnet cost of the System 311 is
low~r than that of the leading
competitive system which offers
neither error detection nor tape
duplication. The Tally system's
transmission speed brings about
lower phone line costs for a gi ven
amount of data. If used only one
hour 'daily, the System 311 will
send over a quarter-million more
words in a month than will its
chief competitor.
(For more information, designate
#46 on the Readers Service Card.)
men

SYSTEM 600 DIAL-O-VERTER

A new "System 600" Dial-oVerter, from Digitronics Corporation, Albertson, N.Y., will make

it possible for intra-corporate
communications to handle both message and data ~raffic at electro~ic
speeds. It is capable of doublIng
a major corporation's communications capacity for handling administrative messages and data intended for computers. Yet, "System 600"
Dial-o-Verter costs no more than the
slower telegraphic network it replaces and affords the corporation
the opportunity of eventually reducing operating costs.
The system accepts both data
and messages on an int~rspersed
basis. The two types of input are
automatically separated. Major components of the new "System 600" at
the switching center include: a
601 Central Control Station operating on private or toll wires; the
602 Tape Terminal consisting of
two or more magnetic tape handlers;
Model 603 Operator's Console which
controls the entire network of the
system: and Model 604 Printer Terminal is available to monitor all
messages. At the remote locations
the Model 605 Paper Tape Terminals
send and receive automatically under direction of the Central Control Station. The basic system
will handle up to 64 remote highspeed terminals. Additional capacity is available.
Richard W. Sonnenfeldt, President of Digitronics Corporation,
pointed out that "System 600", although factory programmed for each
user's system, is the first ever
to eliminate do-it-yourself customer programming required even by the
latest version of all computer controlled systems; it can be operated
simply as soon as it is installed.
Other important advantages were
listed by Mr. Sonnenfeldt including
provision for special handling of
priority messages.
(For more information, designate
# 47 on the Readers Service Card.)

SIXTEEN CHANNEL PAPER TAPE
TO MAGNETIC TAPE CONVERTER

Digi-Data Corporation, Bladensburg, Md., has announced a new t ranslation device from sixteen channel
paper tape directly to IBM compatible computer magnetic tape. This
portable, completely self-contained
unit is capable of accepting standard sixteen hole paper tape gener-·
ated by rain gauges, level gauges~
temperature sensing devices, loggers, etc.
Front panel digital selector
switches permit the insertion of
twelve digits of fixed data to proCOMPUTERS and AUTOMATION for April, 1966

NewsleHer
vide heading information for the
computer. The conversion system
operates at the rate of 17 readings
(68 digits) per second.
The model 1730 can prepare
magnetic tape suitable for computer
entry from standard rain gauge
tapes. It is believed to be the
only paper to magnetic tape translator commercially available which
accepts complete paper tape spools
(such as those made by Fisher &
Porter Company instruments).
(For more information, tlesignate
#48 on the Readers Service Card.)

Input.Output

TELEMETRICS MODEL 8096
PHOTO·ELECTRIC KEYBOARD

Telemetries Division of TMC,
Santa Ana, Calif., recently introduced a new photo-electric keyboard. According to the manufacturer, this solid-state keyboard
is a new approach to meeting the
industry's requirements for a highly reliable, compact and versatile
encoding device. For any key
pressed, a corresponding code is
presented at the output for entry
into a computer, tape punch, information retrieval system, or
other data handling o~ display
equipment.
The keyboard uses long life
lamps which stimulate an arrangement of photo-sensitive devices.
When a key is depressed, a given
arrangement of shutters, unique to
that key, blocks the light path to
a series of photoelectric sensors,
and initiates a delayed clock pulse.

rangement. Quantity of keys per
row and number of rows per keyboard are controlled by the customer's configuration. This button spacing feature and the modular row approach mean that a wide
range of "special" keyboards can
be implemented quickly and economically.
The keyboard is readily adaptable to new applications since
changing of output codes is easily
accomplished by untrained personnel in an hour or two. The simplicity of design provides a device that is lightweight, compact,
easily assembled and tested, at a.
price competitive with similar electromechanical keyboards.
(For more information, designate
#50 on the Readers Service Card.)

THE 'MINITYPER'

A new low cost, high speed
printer, the "Mini typer", has been
developed by Shepard Laboratories,
Inc., Summit, N.J., for use with
analog-to-digital converter output
(Voltage, Current, Frequency), BCD
Counter Output, Telemetry Printout,
Computer flip-flop printout and
similar applications.
It has speeds of 20 lines per
second alpha-numeric or 40 lines
per second numeric, with 1 to 48
characters per line. The Shepard
"mini typer" will accept any input
codes of up to 6 binary. digits in
accordance with customer specifications. Logic levels at any two
voltages, plus or minus, with difference in levels of greater than
3 volts. The "Mini typer" has all
solid state electronics and a selfcontained power supply. It is only
8 3/4" high, and is available as a
table top cabinet or as a 19" rack
mount.
(For more information, designate
#51 on the Readers Service Card.)

CONTINUOUS FORMS
FEED FUNNEL FOR
PAPER SHREDDER

-- Model 8096 Photoelectric Keyboard
Keys are in modular rows and
each row can generate 8 bits plus
a clock pulse. The keys are arranged on 3/4 inch centers in rows
of 13 or less, to any customer arCOMPUTERS and AUTOMATION for April, 1966

Computer. print-outs, single
and multiple continuous forms up
to 18" wide now can be shredded
easily with a new attachment, the
Continuous Forms Feed Funnel, recently developed by Electric Wastebasket Corp. of New York for their
Super Speed Destroyit Paper Shredder. The new attachment makes
p~ssible the security destruction
of any type of continuous form,

fan-fold or roll; it can accommodate
multiple carbon copy forms without
removing staples or Clips.

Where limited space restfictions warrant an orderly method of
forms disposal, the Super Speed
Destroyit, with a capacity of up to
500 Ibs. of shreds per hour, provides rapid on-the-spot destruction. The Super Speed shredder is
available in several models, each
differing only in its respective
shred width of 3/16", 1/16" or
1/32", depending upon the degree
of security required. All models
are driven by a powerful dual drive
5/8 H.P. electric motor and run on
standard 110 A/C current.
(For more information, designate
#49 on the Readers Service Card.)

PORTABLE DIGITAL·RECORDER

The Ralph M. Parsons Company,
Pasadena, Calif., now is producing
a new digital tape recorder to be
known as the DR 1200. This small,
portable device reads and writes
IBM computer compatible tapes at
200, 556, and 800 bits per inch,
with tape speeds up to 120 inches
per second. Recording format is
7 or 9 track data on IBM reels.
The DR 1200 weighs only 45
pounds and operates with only 100
watts of power. Overall dimensions
are 19 in. x 14 in. x 7.5 in. Design is highly flexible and the
recorder can be readily adapted to
meet the requirements for field or
fixed applications throughout a
broad range of environments.
(For more information, designate
#52 on the Readers Service Card.)

HIGH·SPEED DIGITAL PRINTER
FOR DATA·LOGGING MARKET

DI/AN Controls Inc., Boston,
Mass., has introduced the lowest
priced high-speed digital data
printer presently available for
digital data-logging applications.
1\9

Newsletter
The Series "DL" Lister/Printer is
capable of listing 40 lines of numeric data per second and printing
20 lines of alpha-numeric data per
second.
The printer has a simplified
design with only 4 major functional assemblies. An exclusive inkroller exceeds the inking capcity
of 10 standard ribbons and will imprint more than 10,000,000 lines
before replacement is needed. The
individual ballistic impact assembly for each column consists of
only two moving parts. Paper feed
is accomplished by a simple twopart capstan drive which provides
paper motion. Paper input and output bins, which are located on the
front panel, have a capacity of
1000 sheets of 2" x 8~" fan-fold
paper.
Through a variety of performance and interface options the Series "DL" can be economically and
easily adapted to virtually any
digital data source. The Series
"DL" wi 11 accept any binary character code. It has four input data
circuits to choose from, and optional line storage with 2 ~sec
transfer time is available. The
machine operates asynchronously or
synchronously.
(For more information, designate
#53 on the Readers Service Card.)

NEW PRINT-PUNCH
MARKING SYSTEM

A new print-punch marking system has been introduced by Litton
Industries' Kimball Systems division, Belleville, N.J., The new
system makes it possible to machine
process a variety of hopper-fed
tags with pre-punched "size" holes
at the rate of 175 per minute without stopping to change size data.
The Kimball hopper feed system
utilizes a Kimball PM 75 punchmarking machine to eliminate multiple set-ups of punch-mark tags.
The combination of hopper feed and
and pre-punched size information
enables the user to select tags
wi th pre-punched holes and run them
through the Kimball PM 75 punchmarking machine without additional
set-up or processing. This eliminates the possiblility of human
error in punching size information
into tags at each size change.
A special feature enables the
user to punch-mark the exact quantity of tags required by setting
automatic controls. The system

Components

NEW PROCESS OPENS WAY
TO GENERAL APPLICATION OF
INTEGRATED CIRCUITS IN ALL
OF COMMUNICATIONS INDUSTRY

"""
,

provides a fully automatic. and
versatile print-punch markIng
method for the retail industry.
(For more information, designate
#55 on the Readers Service Card.)

SERIES 300 PRINTERS
BY DATAMARK, INC.

A new line of Terminal Printers has been announced by Carl I.
Wassermann, President of Datamark,
Inc., Westbury, L.I., N.Y. The
printer system can be used to pr~nt
data received over a telephone lIne
at its maximum capacity using a
Model 201 or similar Dataphone.
The interface is easily tailored to
operate with any computer. Any
data coding can be accommodated.
Standard horizontal spacing is 10
characters per inch; vertical spacing is 6 and 8 lines per inch. The
paper slew is 3600 lines per minute.
Printers in the new 300 Series
do not use hammer sharing or paper
side-stepping techniques. Printer
specifications are geared to Data
Communications, low cost tape to
printer and card to printer systems and as small scale computer
outP~ts. Worst case print speed
is 300 lines per minute with full
64 character type font. Ninetysix character fonts are available
on special order.
The printer uses nylon or silk
ribbons and accommodates plain or
pre-printed forms, single or multiple carbons, card stock, pressure sensitive, or transfer masters. Frame sizes are available
for 80 and 132 columns.
(For more information, designate
#54 on the Readers Service Card.)

An experimental process that
opens the way for the first time to
general application of integra~ed
circuits throughout the communIcations industry, including radio,
television, radar, microwave, and
other systems, has been announced
by the Radio Corporation of America.
The new process, developed at RCA's
David Sarnoff Research Center,
Princeton, N.J., produces halfdollar-size mosaics made up of hundreds of tiny silicon tiles, each
containing a microelectronic circuit element. The entire assembly
is set in a ceramic matrix that insulates the elements electrically
from one another.
The experimental process was
conceived by Arthur I. Stoller in
a long-range Process Research and
Development program at RCA Laboratories under the direction of C.
Price Smith. It now is undergoing
evaluation by RCA's Electronic Components and Devices organization
at Somerville, N.J. According to
Mr. Smith, present integrated circuits are manufactured on wafers
of silicon in the form of tiny rectangles which are insulated from
one another by one of two techniques.
Both of the techniques are suitable
for the production of low voltage
digital (computer) circuits and a
few types of analog (communications) circuits, he said, but
neither is applicable to the vast
majority of communications circuits
which require higher voltages.
The new process provides an
insulating border around each tile
that can withstand voltages 100
times greater than those which can
be contained by the present techniques, Mr. Smith explained. He
added that the new process requires
no modification to long-established
diffusion techniques for building
microcircuits into tiles, or for
depositing the circuit interconnections in the desired pattern. He
said that the insulating material
used in the new RCA process is a
low-loss ceramic which is not only
resistant to high voltages, but
also is easy to use and produces a
far more rugged circuit mosaic than
can be achieved with the pr-esent
processes.

COMPUTERS and AUTOMATION for April, 19GG

NewsleHer

.to

ELECTRONIC SAFETY DEVICES
MADE BY NEW COMPANY

MASTER NEGATIVES OF GLASS
PRODUCE CIRCUIT BOARDS

A device made by Pacific Data
and Controls, Portland, Ore., preserves signals transmitted down
long lines or over radio circuits
for hours in case lines go down or
power fails. It also is used in
digital computer control systems.
Called a sample and hold amplifier,
it operates on a newall-electronic
principle which eliminates costly
and bulky mechanical arrangements
previously used.

Photoplates made of chemically strengthened glass are being
used as master negatives to produce circuit boards for IBM System/360 computers. Each master is
used to produce many circuit boards
and is remade to a new pattern as
engineering changes are placed in
production •

Model 101 memorizes analog input values for hours after input is
removed. This all electronic solid
state amplifier holds values to better than 1% yet it uses no unwieldy
and expensive electro-mechanical
arrangements. When used with Model
4010 Digital-to-Analog Converter in
a digital control syste~, it provides fail-safe operation when the
digital processor fails or is taken
off-line for a length of time. A
unique capacitance-feedback amplifier performs the required memory
function yet reduces cost substantially over the nearest competing
device on the market.
The Model 101, may be used in
a multitude of applications where
information must be transmitted
from one place to another and loss
of this information would cause a
dangerous condition or costly misoperation. Provision is made for
either manual or automatic operation. Transfer from automatic to
manual and vice versa may be performed without discontinuing
service.
(For more information, designate
#56 on the Readers Service Card.)

COMPU-PSI
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 : 2009:03:20 15:05:58-07:00
Create Date : 2009:03:20 15:05:58-07:00
Metadata Date : 2009:03:20 15:05:58-07:00
Format : application/pdf
Document ID : uuid:efd5b6e1-705d-4949-b0b8-ff9d3caf8cf2
Instance ID : uuid:61a9873f-8f9b-4301-ac5a-d6a68a91f0d8
Page Layout : SinglePage
Page Mode : UseNone
Page Count : 64

EXIF Metadata provided by EXIF.tools

196604

196604 196604

Navigation menu

Versions of this User Manual:

Views

Navigation