196011 196011

User Manual: 196011

Open the PDF directly: View PDF PDF.
Page Count: 48

Open PDF In BrowserView PDF
What Business Needs Most from Manufacturers of
Electronic Data Processors
Undetected Errors in 5-Unit Code Transmission
and Their Elimination
Automation - Its Evolution and Future Direction


Applied Mathematicians
and Computer Programmers
to Participate in
Advanced Space Programs
The Missile and Space Vehicle Department of
General Electric - a recognized leader in the
development of instrumented re-entry vehicles is now pursuing a number of even more advanced
space programs. Basic to progress in these programs is the solution of a diversity of interesting
mathematical problems. These include trajectory
and navigation studies and analysis of flight
telemetry data and space communications.

The applied mathematics group provides consulting services to our entire engineering staff. Salary
and professional growth are given corresponding
emphasis. Desirable experience and background
would include strong analytical ability, extensive
knowledge of advanced techniques in numerical
analysis for computers, and experience in mathematical investigations on advanced engineering
programs. MS or PhD in Mathematics or Theoretical Physics required.

Diversified Positions for:

As Senior Programmers at the Missile and Space·
Vehicle Department you will have all the advantages of an extensive computer facility which
centers around an IBM 7090. The work covers
analysis and programming for technical data systems, flight test data systems and advanced space
programs. Requirements include ability to direct
junior programmers, a BS or advanced degree,
minimum of 2 years experience on a large scale,
binary computer.

BSEE, Physics or Math degree required. Will
plan sequence of computer operation, determine
the circuitry for engineering problems, set up and
operate computer.

The work is in a growing analog facility which
includes Electronic Associates and Reeves Analog
Equipment, a combined Analog-Digital Facility
and a passive element analog computer.
For further information regarding opportunities
here. write Mr. D. G. Curley, Div. 21-MK.


3198 Chestnut Street, Philadelphia 4, Pat


COMPUTERS and AUTOMATION for November, 1960

-- .... - ... ----- ...


is the Philco mathematical language
compiler which operates in Conjunction With
TAC, the Phil Translator .Assembler_Com_
Piler · · . a Powerful, computer-oriented lan_
gtlage Witb extensive library features.
ALTAC is also compatible With FORTRAN
II*. · . permitting tboUsands ofeJriSting programs
to be run on the Philco 2000 ComPuter Witbout
change. It produces a fast, efficient running


program. T AC langtlage inserts may be included
at Will in ALTAC langtlage programs.
ALTAC is now in full operation at CUstomer
Philco automatic Progranumn techniques
enable you to learn and Use the gPhilco 2000
faster. For more COmplete inf _



Volume 9
Number 11

Assistant Editor
Assistant Editor
Assistant Editor

439 So. Western Ave. OR 815 Washington St.
Los Angeles 5, Calif.
Newtonville 60, Mass.
DUnkirk 7-8135
DEcatur 2-5453

Los Angeles 5
439 So. Western Ave.
DUnkirk 7-8135
Chicago 10
18 E. Division St.
San Francisco 5
605 Market St.

MIchigan 2-0778
YUkon 2-3954


439 So. Western Ave. OR 815 Washington St.
Los Angeles 5, Calif.
Newtonville 60, Mass.
DUnkirk 7-8135
DEcatur 2-5453
COMPUTERS and AUTOMATION is published 13 times
a year (monthly except two issues in June) at 815 Washington St., Newtonville 60, Mass., by Berkeley Enterprises,
Inc. Printed in U.S.A.
SUBSCRIPTION RATES: (United States) $7.50 for 1
year, $14.50 for 2 years; (Canada) $8.00 for 1 year, $15.50
for 2 years~ (Foreign) $8.50 for 1 year, $16.50 for 2 years.
Address all Editorial and Subscription Mail to Berkeley
Enterprises, Inc., 815 Washington St., Newtonville 60,
Office at Boston, Mass.
POSTMASTER: Please send all Forms 3579 to Berkeley
Enterprises, Inc., 815 Washington St., Newtonville 60,
Copyright, 1960, by Berkeley Enterprises, Inc.
CHANGE of ADDRESS: If your address changes, please
send us both your new address and your old address (as
it appears on the magazine address imprint), and allow
three \veeks for the change to be made.


September 1951

Vol. 9, No. lIB

News of Computers and Data Processors:
inserted between pages 8 and 9
and between pages 24 and 25

Digital Computer Demonstrator


What Business Needs Most from Manufacturers of Electronic Data Processors, BENJAMIN CONWAY .
Undetected Errors in 5-Unit Code Transmission and Their
Elimination, JAMES F. HOLMES
Automation - Its Evolution and Future Direction (Part
Daedalus: "Complete-The-Square" Computer - Correction, S. F. GRISOFF
Four Prefixes for Sizes, FRANK LEARY .
First International Conference on Automatic Control:
The Social Responsibilities of Computer People - Reports and Discussions at the Meeting of the Association
for Computing Machinery, August, 1960 (Part 2)
Changes in Commercial Computers Anticipated in Coming Years
Calendar of Coming Events
Books and Other Publications, M. M. BERLIN
Essential Special Terms in Computers and Data Processing - Suggested List, and Definitions (Part 3)
New Patents, R. R. SKOLNICK




Advertising Index
· 30
Back Copies
· 28
Computer Directory
· 26
Glossary of Computer Terms
· 29
· 27
see August, page 2
Reference and Survey Information
Statement of Ownership and Management
Who's Who Entry Form
COMPUTERS and AUTOMATION for November, 1960


At North American Aviation's Rocketdyne division, the recording of rocket engine performance on static stands, calls
for the use of only the most reliable of instrumentation tapes,
like Soundcraft. Why? Because Soundcraft Instrumentation
Tapes combine exact physical and electrical properties so
necessary to assure absolute reliability of performance-as a
direct result of being manufactured in the world's most
modern tape plant under the most advanced quality control
In short, experience has proven that Soundcraft Instrumentation Tapes work best in recording of critical rocket testing
operations. Why not let precision-made, error-free Soundcraft Instrumentation Tapes go to work for you? Complete
literature on request.

COMPUTERS and AUTOMATION for November, 1960

Great Pasture Rd., Danbury, Conn .• Chicago: 28 E. Jackson Blvd. los Angeles: 342 N. laBrea 8 Toronto: 700 Weston Rd. cO


What Business Needs Most Fronl
Manufacturers of Electronic
Data Processors
Benjamin Conway
Director, Management Advisory Services
Price Waterhouse & Co.
New York, N.Y.

(Reprinted with permission from the N.A.A. BttlietinJ August, 1960, vol. 41, no. 12,
published by the National Association of Accountants,
505 Park Ave., New York 22, N.Y.)

It mIght be well to consider briefly the requests businessmen are now making of equipment manufacturers before
proceeding to what might be required in the near future.
Some of these points may seem trivial or obvious, but they
are very real to present and potential users of electronic
equipment and, although it may seem that we are severely
criticizing business machine manufacturers, this is not true.
The manufacturers have done a fine job in their equipment
design and improvements, in their research programs, and
in their efforts to help prospective users. The points being
made now are to help the manufacturer raise his already
high standard of performance.
User Needs as Related to Present Equipment
First and perhaps foremost, business users would like to
have accurate specifications on the proposed system. Most
users in the past were reasonably sympathetic with the manufacturers' design and production difficulties and would
accept small variations from the original specifications.
However, when this particular piece of equipment or that
special feature failed to materialize or when design changes
necessitated major increases in cost, industry users began to
balk. It seems only fair that, when a user has sunk several
hundred thousand dollars in preparing for a computing system, he should receive the system for which he has prepared.
An important point and one which may have stopped
many prospective users almost before they started is the
high initial cost of an electronic data processing application. The costs of systems analysis, systems design and coding, site preparation, "debugging," and systems testing add
up to a very substantial sum. When to this is added the
rental costs for the equipment when the system in producing little useful work during the low initial loading and
the parallel operation periods, the total cost gets to be
very high indeed. Many people feel that the manufacturers
should attempt to revise their rental structure so that only
low rental costs are incurred at these early stages and that
the rentals could then rise above current figures as the

system becomes more and more productive. Perhaps this
could be achieved by reducing initial costs and increasing
second and third shift rentals. In any case, no matter how
it is arranged, a new approach to rental structure would be
To further reduce initial costs, I believe that the manufacturer should attempt to introduce programming aids at
a very early point in the installation. Experience in the past
has tended to show that service routines, automatic coding
routines and so on, have not been available when the effort
first got under way. Because of this, most user programming
groups have had to work in machine coding and have had
to develop their own service and housekeeping routines.
We know that the situation has been greatly improved and
that most manufacturers have large programming research
staffs. However, there is still a feeling that these staffs do
not always produce routines which business feels it needs.
This lack of communication can be overcome by the better
utilization of users groups; the manufacturers should encourage the setting up of a type of liaison with these groups
so that their programming efforts will yield practical results.
Another point at which the manufacturer and the user
of electronic data processing equipment have not always
seen eye-to-eye in the past concerns equipment delivery
dates. This has tended to work in one of two ways. The
first has been that the equipment delivery has been delayed.
When this happens, the user is forced to keep an expensive
programming group sitting around idly. Alternatively,
the user has often hopelessly underestimated the time required for programming and conversion. In this case the
system is installed before it can be used effectively, another
costly procedure. Since the user has often relied heavily
upon the manufacturer's representative for technical assistance in setting up his timetable, the user, perhaps illogically,
blames the manufacturer for this state of affairs. The
answer from the manufacturer's viewpoint, of course, is
that he should, through his technical representatives, assist
the user in drawing up a realistic program timetable and
also that he should set reasonable delivery dates for himself.
COMPUTERS and AUTOMATION for November, 1960

Still a further requirement that industry would place
upon the manufacturer i:; that of providing adequate breakdown coverage. A computing system which is not functioning is the nightmare with which executives often have to
live. Their lot would be considerably improved if they
knew that time was always available to them locally on
similar equipment. If this ideal cannot be achieved, the
manufacturer should at least provide standby equipment of
his own at some reasonably located point. It should never
be necessary, say, for a user in San Francisco to have to
fly his personnel and tapes to New York in order to keep
his paperwork processes functioning.
Among yet other important matters on which the manufacturer should supply information are adequate details for
site preparation and air conditioning. I do not know why
manufacturers should be so modest in this respect, but I
have knowledge of at least two installations for which the
space recommended by the manufacturer had to be considerably increased to provide good working conditions. It
is much easier to allow extra space at the planning stage
than to have to create it when the program is well under
In the areas of programming training and aids, the manufacturer has his greatest non-machine responsibilities. The
manufacturer's representative is expected to take a group
of inexperienced people and train them in an art which
is, to say the least, complex. To do this, the representative,
himself, must be fully trained, both from the theoretical and
practical viewpoints. He must also be aware of the principles of teaching, since knowledge and the communication
of that knowledge are two entirely different things. In
addition, the representative must be backed by adequate
training material such as programming manuals, programming sheets, templates for charting, visual displays and so
on. It is to the manufacturer's advantage that this material
be well prepared, since nothing can destroy the confidence
of a beginner more than to see a programming manual full
of errors and corrections.
As a good installation is almost as vital to the manufacturer as to the user, it is also desirable that the training
courses be adequately supervised and the participants be
tested at its close. Also, where required, the manufacturer's
representative should be prepared to comment upon the
suitability of any individual where, during the training
course, it becomes apparent that such an individual may
not fit into a programming group. Characteristics such as
over-aggressiveness, disinclination to work and inability to
work under pressure may show up during a prolonged
training period, even when the individual concerned has
passed all the theoretical testing procedures. These characteristics, which may cause trouble in a working group,
should be commented upon.
Very importantly, routines should be available which will
assist the user in program "debugging" and, fairly obviously,
adequate machine time should be made available to a prospective user in order that he can "debug" and systems-test
his initial programs prior to delivery. If the manufacturer
feels that he cannot increase present "debugging" time free
of charge, he should endeavor to work out some arrangement by which the extra time would be gr?nted and paid
back at a later date by the manufacturer taking time on
the user's equipment.
COMPUTERS and AUTOMATION for November, 1960

Wherever possible, the manufacturer's technical representative or representatives assisting a user should not be
changed at least until the initial programs are operating
satisfactorily. Even if the replacement is of equal calibre to
the replaced man it can be a time-consuming business to
bring him up to the required state of knowledge. If replacement is unavoidable, the user will obviously take a
reasonable point of view but, if he believes that the company is being used as a training ground for the manufacturer, unpleasantness will probably arise.
The manufacturer must also ensure that the computer is
adequately maintained. This requires not only the services
of an expert maintenance man or team but also the establishing of an adequate source of parts and supplies on the user's
premises to cover all but major breakdowns. It also requires
the setting up of a preventive maintenance program at times
suitable to both the user and the manufacturer.

And Future Requirements

The foregoing material has dealt with present requirements and existing equipment. Now let us turn to some
of the requirements that industry will be making in the
near future. Let us consider first companies now uncommitted, the potential users. For the companies in this category, cost is probably of major importance since most large
companies, or companies in which money for new projects
is fairly reasonably available, will already, in one way or
another, have made their approach to electronic data processing. These presently potential users, then, would be intensely interested in cheaper equipment than is now available. While it is true that modern systems offer far more
for the computing dollar than earlier machines, this secondgeneration equipment has not decreased the basic cost at
the lower end of the scale. Rather, the tendency has been
more to introduce a new category of equipment with medium and medium-large costs which can perform as much
work as the previous large-scale systems. This is evidently
a step in the right direction and should be continued
throughout the entire gamut of electronic machines. Wh?t
is needed in addition, however, is that there should be a
class of machines which will do the work now done by the
present medium and small-scale tape-and-card systems at
much lower total costs. With modern computer logical
designs and extensive use of transistors, it would seem that
this should not be too difficult for manufacturers to achieve.
If this is not so, the manufacturers should make announcements to this effect. There is a large group of potential users
awaiting such equipment.
Through talking to representatives of many users, we feel
that this reduction in COSt is a much more desirable trend
than aiming at increasing machine capacity and capability.
Responsible people in industry feel that it will be many
years before they have learned to use the capabilities of
present equipment. Hence, they remain comparatively unimpressed by the announcements of machines which will
operate in micro-micro-seconds instead of in micro-seconds.
These people feel that most of the features that they require in computing systems are presently available, although
not always in the same systems. A consolidation of such
features as core memory, parallel rather than serial transfer
of data within the computer, adequate internal control
checks, faster input-output equipment, a greater degree of

slmultaneity between input, processing and output, and the
ability to perform multiple programming seems to be much
more important than a fantastic increase in the speed of
internal processing.
A major cost associated with electronic data processing
when operating in a decentralized company is that of communications. This is an area in which industry would like
to see some major improvements. At present wire communication networks are tied up by the relatively slow speed
of transmitting information and the very slow speed of
receiving the information. These networks would be
speeded greatly if a paper tape or punched card transmission could be recorded directly to magnetic tape, and could
be speeded even more if data were transmitted on a magnetic tape to magnetic tape basis. These tapes should, of
course, be compatible with the computing system. Among
other requirements of a communications system, would be
the ability to perform some editing of the information while
it is being transmitted and better methods of error detection
and correction than are currently available.
It was said above that most of the features of electronic
data processing equipment which are considered desirable
by industry can be found in one or more of the modern
computers. There is one area, however, in which many
users feel that much can be done. I am referring to the
area of random access. If a random access system of several
million character capacity could be developed at approximately the same total system cost as with mass magnetic
tape storage, then present methods of processing could be
completely changed and many applications not presently
suitable for electronic data processing would fall easily into

The storage of information on numerous reels of magnetic tape necessitates multiple machine runs, a great deal
of lost time in having to pass the same tapes through the
system many times, and extra time spent in sorting and
collating. It creates difficulties in interrogation of information contained in the tape files and may, in cases in which
these interrogations need to be answered immediately, require the addition of expensive off-line interrogation units
or voluminous print-outs in order that normal processing
should not be unduly disturbed. Large random access systems at the level of cost I have indicated would obviate the
necessity for all this and would, therefore, greatly simplify
electronic data processing and, at the same time, increase
the usefulness of the systems. There would be no need to
sort incoming items; they could be handled in the order in
which they arrive. There would be no need for multiple
master files or numerous machine runs and all the processing, including file maintenance, would be performed in
a single run. I do not know what the manufacturers have
"in the mill" for the development of truly large, truly cheap
random access systems, but it does seem to me, and to
others, that in the area of machine design this could prove
to be the most fruitful from the users' viewpoint.
The area which would seem to promise most, outside
of the actual equipment, has been mentioned briefly earlier.
It is automatic coding. At this point, I want to broaden the
area to include a universal machine language and the development of systems packages. It just seems unreasonable
to us that different groups of people have to go through
the same gropings, through the efforts of systems analysis

and design, through the drudgery of coding without being
able to take full advantage of the experiences of others in
similar work. When we consider, for example, the many
public utilities which, completely independently and often
in a most amateurish manner, have developed systems for
customer accounting functions, the waste of time and money
involved becomes incredible. Customer accounting may
vary from one utility to another but the essential character
of the job remains the same. A generalized program developed by a professional group which could then have been
adapted to the particular needs of anyone utility could
have resulted in the savings to each user of many thousands
of dollars, and the savings to the industry as a whole could
perhaps run into the millions. We are aware that, as a
practical matter, utilities which have had to struggle along
on their own might not be too interested in joining a combined industry group now, but we feel that they could be
so persuaded if it could be effectively demonstrated to them
that such participation would benefit them in the development of new applications and, just as importantly, in the
revisions of the programming of existing applications. We
believe, however, that the impetus for this must come from
the machine manufacturers. We are aware that some manufacturers have industry specialists who do help to carry the
experiences of one company to another, but we feel that
this falls short of the goals which can be attained.
By and large manufacturers have tended to shy away from
the broad "language" area as being outside their sphere of
responsibility or as being impracticable of development
by them but we believe that this is untrue. We feel that
there is sufficient work of a like nature within each industry
or group of industries to make the development of the
systems approach a worthwhile and important endeavor.
We feel, furthermore, that manufacturers who do develop
the systems approach will have a tremendous competitive
advantage over others. Costs of this could perhaps be borne
by the manufacturer and industry group.
These then we feel are the three basic future requirements for manufacturers:
1. To develop cheaper equipment at all levels of

equipment capabilities, especially the medium and
small scale systems.
2. To develop random access systems which are gigantic in comparison to modern equipment and at a
price not far out of line with tape storage.
3. To develop the systems package encompassing both
the systems and the machine rather than the machine alone.
Mutual Interests of Manufacturers and Users
Let us not forget, however, that industry feels that these
are additional requirements on the manufacturer and that he
should continue to supply and improve upon these services.
We know of many instances in which machine selection has
been made on the basis of service supplied in the past
rather than on the characteristics of the machine in question.
As acknowledged at the start, what has been said on these
pages is not meant to be a criticism of manufacturers. They
have performed well in the past, but much remains to be
done in the future and both manufacturer and user will
benefit from further, substantial progress.
COMPUTERS and AUTOMATION for November, 1960





sells the

[m~~[m~~ W~~ ~@~]~~
in the multi-billion
dollar computer and
data processing


Perhaps the Key Reason Why



Is The Only Monthly Trade Publication Serving Both Builders And
Buyers In This Mushrooming MultiBillion Dollar Industry

Is That It Was Founded And Is
Edited By Edmund C. BerkeleyPioneer Computer Mathematician



• Author of "Giant Brains or Machines
that Think", John Wiley and Sons, 1949

Addressograph Multigraph, Aladdin Electronics, Alwac,
American Machine and Foundry, Ampex, Arnold
Engineering, Automatic Electric, Beckman Instruments,
Bendix Aviation, Bristol Company, Bryant Gage and
Spindle, Burroughs, Clary Multiplier, Computer Control,
Datamatic, Daystrom Systems, Electronic Associates,
Epsco, Ferranti Electric, Ford Instrument, General
Controls, General Electric, General Kinetics, Goodyear,
Haller Raymond and Brown, International Business
Machines, Jet Propulsion Labs, Lear, Leeds & Northrup,.
Link Aviation, Litton Industries, Melpar, MinneapolisHoneywell, National Cash Register, Northrop Aircraft,
Packard Bell, Panellit, George A. Philbrick Researches,
Phi1co, Potter Instrument, Ramo-Wooldridge, Raytheon
Mfg., Reeves Instrument, Royal McBee, Sperry-Rand,
Stromberg Carlson, Sylvania Electric Products, Telemeter
Magnetics, Texas Instrument, Westinghouse . . .

• Co-author of "Computers Their
Operation and Applications", Reinhold
Publishing Co., 1956
• Author of "Symbolic Logic and Intelligent
Machines", Reinhold Publishing Co., 1959
• Maker and Designer of small computing
machines including the BrainiaC® electric
brain construction k~ts, Simon (miniature
complete automatic digital computer),
Relay Moe (tit-tat-toe playing machine
pictured in Life Magazine, March 19,
1956), etc.
• Fellow of the Society of Actuaries;
Harvard 1930 A.B., summa cum laude in
mathematics; author of many articles and
papers in New York Times Magazine,
Scientific American, Record of the
American Institute of Actuaries, Journal
of Symbolic Logic, etc.
• Entered computer field in 1938.
• Founded "Computers and Automation"
in 1951.

The Computer & Data Processing
Field Is Huge!
When Edmund C. Berkeley founded

1951, the industry was just starting to flex
its muscles! Now it is estimated that the

computer and data processing market will
be "between five billion and nine billion
dollars over the next five years".

Computer BUILDER Subscribers include:


Computer BUYER Subscribers include:
Arthur Anderson & Company, Blue Cross, Ehasco
Services, Price Waterhouse, Bank of America, First
National Bank of Boston, Bank of New York, ChaseManhattan, First National Bank of Atlanta, Provident
Trust, Equitable Life Assurance Society, Great Western,
Metropolitan Life, Penn. Mutual, U.S. Life, 'Battelle
Memorial Institute, Brookhaven National Laboratory,
Bureau of the Census, National Bureau of Standards,
Los Alamos Scientific Laboratory, National Weather
Record Center, Port of New York Authority, Railroad
Retirement Board, Rand, U.S. Air Force, U.S. Security
Service, U.S. Army Tank Ordnance Center, U.S. Navy,
David Taylor Model Basin, University of California,
Kansas State Gollege, Mass. Institute of Technology,
San Diego State College, Wayne State University,
University of Wisconsin, Ford Motor, General Motors,
General Tire and Rubber, American Cyanamid, Eastman
Kodak, Owens Corning Fiberglass, Monsanto Chemical,
Allis Chalmers, Cutler Hammer, Diehl Mfg., Western
Electric, Westinghouse, Calloway Mills, Kellogg, Briggs
and Stratton, United Shoe, Atlantic Refining, Esso
Research, Pan American, Sun Oil, Texas Eastern
Transmission, Bethlehem Steel, Continental Can, U.S.
Steel, American Tel. and Tel., Arizona Public Service,
Bell Telephone Labs., Consolidated Edison, Florida
Power, Pacific Gas and Electric, Johns Mansville,
Boeing Airplane, Convair, Douglas Aircraft, Grumman,
Hughes Aircraft, Lockheed, North American Aviation,
Reaction Motors, American Airlines, Hawaiian Airlines,
Pennsylvania Railroad.

Companies and Products Advertised in
(Below is a partial list ranging from
June 1959 to September 1960)

General Electric
Knolls Atomic Power Lab.
Semiconductor Products Dept.
Missile and Space Vehicle Dept.
Apparatus Sales Office
Heavy Military Electronics Dept.
C. P. (lore
Bendix Corp.
Honeywell Dotamatic
Aeronutronic Systems
Royal McBee
System Development Corp.
Arnold Engineering
Audio Devices
Phil co
Sylvania Electric
Lockheed Aircraft
Technical Operations
Space Technology Lobs
C-E-I-R, Inc.
DilAn Controls
NJE Corp.
Computer Control Co.
Internotional Business Machines
Boeing Aircraft
Reeves Soundcraft
National Cash Register Co.
Packard Bell
Raytheon Mfg. Co.
Potier Instrument
John Wiley & Sons
John Diebold
Chrysler Corp.
Republic Aviation
Aeroiet General
Reinhold Publishing Corp.
Wheeler Fairchild, Inc.·
Douglas Aircraft
Ferranti Electric, Inc.
Mitre Corp.
Sandia Corp.
Hughes Aircraft, Co.
Computer Systems, Inc.
Texas Instruments Inc.
Norden Dlv., United Aircraft
Bell Aircraft Corp.
Information Systems, Inc.
Burr & Co.
Dynacor, Inc.
Clary Corp.
Library of Science
F. W. Sickles
Adams Assoc., Inc.
Cambridge Communications Corp.
Electralab Printed Elec. Corp.

Product AdverHsed
Computers, Components, Photoreader,
Employment Opportunities
Transistors, Cores, Employment
Opportunities, Cells, memory planes,
memory cores, systems
Pulse Transformers, Delay Lines
Tubes, Capacitors, Consulting Service
Employment Opportunities
Employment Opportunities
Employment Opportunities
Power Supplies
Employment Opportunities
Tape, Tape Handlers
Relays, Wiring Services, Stepping
Switches, Insulation
Digital Computing Systems, Paper
Tape Reader, Couplers, Employment
Computers, Employment Opportunities,
Data Processing System, Rentals
Employment Opportunities
Capacitors, Employment Opportunities,
Computers, Computer Services
Employment Opportunities
Cores, Magnets
Employment Opportunities
Computers, Data Processing Systems,
Employment Opportunities
Employment Opportunities
Employment Opportunities
Employment Opportunities
Employment Opportunities
Employment Opportunities, Consulting
Cores, Buffers, Interloks,
Logic Squares
Power Supplies
Memory, Modules, K Pacs
Employment Opportunities
Employment Opportunities
Tape, Kit to Check Recording
Computers, Circuitry, Memories,
Employment Opportunities
Indicator Lights
Computers and Data Processing
Systems, Components, Alphanumeric
Consulting Services
Employment Opportunities
Employment Opportunities
Employment Opportunities
Magnetic Storage Drums
Employment Opportunities
Delay Lines
Employment Opportunities
Employment Opportunities
Memo-Scope Oscilloscope
Employment Opportunities
Employment Opportunities
Employment Opportunities
Employment Opportunities
Equity Capital
Bobbin Cores
Delay Lines
Adding and Posting Machines
Electronic Data Processing,
Consulting Services
Abstracts on Cards
Printed Wiring, Circuit Assemblies


Will Be Read By Even More People Who Count
During 1961, we plan that the circulation of COMPUTERS
and AUTOMATION will reach at least 5000 copies, as
compared with 2600 subscribers in 1957, 3700 in 1958,
and 4000 in 1959 and 1960. A continuing circulation
campaign promises to add numbers of new subscribers to
our audience. All new subscriptions are regularly paid for at
$7.50 a year, assuring a top-quality as well as a top-quantity
computer-minded audience. About 500 key decision-makers
working at computer builders and at computer buyers have
been included on a controlled basis in order to assure a very
important marketing audience.






Ed itorial Coverage is Expand ing, Too
includes new articles, new reference information, and new
ideas, to make it as complete, informative, ~nd stimulating
to computer people as we can make it. Greater emphasis
than ever before is being placed on news of computers and
data automation . . . new surveys, new applications, new
products, new devices, new ideas about computers and data
processors ... everything we can think of to help our readers
do their jobs better, more quickly, and more profitably.
In February 1960, COMPUTERS and AUTOMATION
went to 24 "issues" a year - a dozen regular letterpress
issues closing the 1st of the previous month, and a dozen
photo-offset issues "News of Computers and Data Processors: Across the Editor's Desk" closing the 15th of the
previous month. The two "issues" are published together
at the beginning of the current month, except that in
June they are published separately.

CompUTfRS Hno AUTomHTlon
One Time

Six Times
5% Discount

Twelve Times
10% Discount

1 page




% page




% page




% page




% page








Middle Atlantic States
535 Fifth Ave.
New York 17, N.Y.
MUrray Hill 2-4194
Washington 6, D.C.
1519 Connecticut Ave.
COlumbia 5-9727


San Francisco 5
605 Market St.

YUkon 2-3954

Los Angeles 5
439 S. Western Ave.

DUnkirk 7-8135

Berkeley Ent6q)rises, Inc.
815 Washington St., Newtonville 60, Mass.
DEcatur 2·5453 or 2-3928






Volume 9
Number lIB







NOVEMBER 2, 1960

September 1951

Published by Berkeley Enterprises, Inc., 815 Washington St., Newtonville 60, Mass.

Burroughs Corp.
Detroit 32, Mich.
An electronlc teleprinter, which prints
at the rate of 50 words per-second or 3000
words per minute, is at this company. Some
of the machines are currently being employed
in the U. S. Air Force Weather Observation
and Forecasting system, and Global Communications program, and in projects for other
branches of the government. Burroughs will
ship some 50 of these machines by the end of

The teleprinter records alpha-numeric or
symbolic characters at rates up to 300 characters per second on roll paper 72 characters
wide. It can be used as a readout from electronic computers, or operated from input signals from polka dot, punched or magnetic tapes,
or telegraphic signals. It has capabilities
of speeds up to 1000 characters per second.

Printing by the electrostatic teleprinter
is accomplished by a row of solid-state electronic print heads in matrix form that charge
paper with electrostatic charac~er images.
These images pick up powered ink as the paper
passes through an ink bath and are fixed in
hard, permanent copy when the paper passes
over heated pressure rollers. The ordinary
hammering type faces, typical of some impact
printers, are eliminated.
Since the paper never touches the print
heads, and the only moving parts are associated with the paper advance mechanism, the
life and maintenance of the teleprinter is
enhanced. In long distance communications
and data processing systems utilizing existing local and long distance telephone lines,
the high speed printer minimizes circuit time
and operational costs.

COMPUTERS and AUTOMATION for November, 1960




International Business Machines Corp.
Data Proces~ing Div.
White Pl.ains, N.Y.
An optical reader of numbers and marks,
which reads at the rate of 480 characters per
second, has been developed by this company.
The machine is identified as the IBM 1418
Reader. As many as 400 documents a minute may
be read. The printed data is automatically
translated into machine language for direct
input to an IBM 1401 computer.
The 1418 reads numbers printed ten characters to the inch in a standard IBM type by
407, 408, or 409 accounting machines, the
1'403 printer, or an electric typewriter. It
can also read numbers in the elongated 407
type s~yle, commonly used by credit· card imprinters, which is seven characters to the
inch. In addition, the 1418 can be equipped
for mark-reading -- in which 'vertical markings made with ordinary pencil or dark 'inks
represent specific information determined by
the format of the,document.
The optical character reader handles
documents of various sizes and thicknesses.
Forms can be anywhere from 5 7/8" to 8 3/4"
wide by 2 3/4" to 3 2/3" high. Their thickness can vary from that of bond paper to IBM
card stock. Any printed documents within
these dimensions -- premium notices, imprinted
charge slips, telephone bills, tax notices,
coupons, and continuous-card forms, for example -- can be used as direct input to the
14i8-equipped 1401 system.


As the documents feed into the 1418, they
are separated and aligned, one document at a
time. Each form then feeds individually onto
a revolving drum, where it is held flat by a
vacuum. As the drum revolves, the complete
surface of the form passes under a lens system.
The standard machine contains one optical
reading station, which will read a single type
style from any line on the document. Available
optionally are either a second reading station
or a mark-reading station. With the second
reaping station, two lines can be read in one
pass of a document through the machine. These
lines can be in the same or different type
The 1418 employs a scanning method which
recognizes a practical range of print quality.
The characteristics by which each number is
identified are contained within the solidstate circuitry of the reader. The light
image of a character is converted into electrical impulses which are compared and matched
with internal logic patterns. Numerical characters are individually recog~ized in this way
and transferred one at a time to the magnetic
core storage of the 1401 system.
Once in storage, the data from the printed
forms can be processed and the results produced
as punched cards, magnetic tape, or printed reports by the 1401. The forms from which the
data were read are fed by the 1418 into its
appropriate sorter pockets.

COMPUTERS and AUTOMATION for November, 1960

International Business Machines Corp •.
Data Processing Div.
White Plains, N.Y.
An electronic "law library" which readily
locates legal information was demonstrated to
members of the American Bar Association at
their recent convention. An IBM 650 data
processing system provided in minutes facts
that would have taken many hours to find by
conventional methods.
The electronic system was able to search
for, find and print out -- upon request -the full text of laws from various states on
health and hospitals.
The data processing unit had stored on
magnetic tape the full text of Pennsylvania
statutes on public health and all statutory
sections of the 1200-page Hospital Law Manual.
This was done by first recording the statutes
on punched cards and then transferring them
-- via the 650 computer and an attached card
reader -- to the tape in the form of magnetized bits.
Anyone of the five tape units in the
system can be called into operation through
instructions previously stored in the computer's magnetic drum memory. The magnetic
drum serves as storage for both the program
of instructions, which controls the operations
performed on the data, and those portions of
the data being worked on. The magnetic tape,
which can be read by the computer at the rate
of 15,000 characters per second, serves as a
permanent storage for the statutory information. This information can be called into the
computer for processing by the program of instructions.
On each 2400-foot reel. of tape it is
possible to record more than 2,000 pages of
running English text. This means that the
equivalent of a 2,000-page legal tome will take
up less than I 1/2 inches of shelf space, the
width of the reel in its canister.
After the text of the statutes was placed
on tape, the computer -- acting on instructions from the magnetic drum -- analyzed the
text and constructed an alphabetical list of
the total vocabulary (eliminating common
words such as "and," "the," "or," etc.).
The computer associated with every noncommon word the identification number of the
statute or statutes in which it appeared.
The list of words and associated statute numbers is the information which the computer
processes in looking for an answer to a query.
COMPUTERS and AUTOMATION for November, 1960

To find the statutory section relevant
to a legal issue, an inquiry for the computer
is framed. The inquiry consists of words, or
groups of words, which are expected to be
found within the desired statutes.
Receiving the inquiry via the card
reader, the computer searches the vocabulary
list which has been placed, along with the
statute identification numbers, on magnetic
tape. The statute identification numbers
associated with each of the selected inquiry
words are compared by the computer. If the
same number is found in all of the selected
word records, then this indicates that the
statute contains all the selected words
and has the information being sought.
For example, an attorney might want to
know what statutes in the Hospi tal Law ilianual
touch on the taxation of charitable hospitals.
An inquiry would be framed -- it might contain the words "charitable," "taxation,"
"exemption," "hospital" -- and given to the
machine. The numbers of all statutes containing all these words would be listed, and,
if desired, the statutes would be printed out
in full.
The Philadelphia Saving Fund Society
Philadelphia, Penna.
This bank has installed an IBM RAMAC
data processing equipment, which will handle
accounting on 720,000 of the bank's savings
accounts (largest number in any U.S. savings
bank). Deposits are more than a billion
The computer already has currently in
its memory all the facts and figures on some
400,000 of the bank's savings accounts.
In addition to deposit accounting, the
machine will shortly do the accounting on
more than 25,000 of the bank's mortgage loans,
calculate the payroll for the bank's nearly
750 employees, and handle many other accounting jobs now done with conventional punchcard equipment.
The machine was tested extensively for
several weeks. As the equipment and procedures became ready, va,ri.ous groups of accounts
were switched over to the machine from conventional punch-card equipment. This will
continue until most of the bank's accounting
procedures are automated, the culmination of
several years' planning and effort.


Processing" - Gomer H. Redmond and
Dennis E. Mulvihill, Chrysler Corp.

DECEMBER 13-15, 1960, NEW YORK -- PROGRAi.l
The 1960 Eastern Joint Computer Conference will take place in New York at the Hotel
New Yorker and Manhattan Center, Tuesday,
December 13 to 1'hursday, December 15. Following is the program:

Tuesday, December 13


Wednesday, December 14

"High Speed Pri nter and Plotter" Frank T. Innes, Briggs Assoc., Inc.
"A Description of the IBM 7074 System"
- R. R. Bender, O. T. Doody and P. N.
Stoughton, IBM Poughkeepsie
"The RCA 601 System" - D. L. Nettleton
and K. K. Kozarsky, RCA
"Associative Self-Sorting Memory" R. R. Seeber, Jr., IBM, Poughkeepsie
"UNIVAC - RANDEX II - Random Access
Data Storage System"- G. J. Axel,
Remington Rand
"Hot-Wire Anemometer Paper Tape Reader"
- John H. Jory, Soroban Engineering, Inc.



9:30 - 11:40 A.M.


Opening Remarks: N. Rochester, Conference Chairman, IBM and E. C. Kubie,
Program Chairman, CUC
"A Logical Machine for Measuring Problem
Solving Ability" - Charles R. Langmuir,
Psycholoyical Corporation
"A Method of Voice Communication With a
Digital Computer" - S. R. Petrick and
H. M. Willett, Air Force Cambridge
Res. Labs.
"FILTER - A Topological Pattern Separation Computer Program" - Daphne Innes,
Lawrence Radiation Lab.
"Redundancy Exploi tat ion in the Computer
Solution of Double-Crostics" - Edwin
S. Spiegelthal, Consultant





Tuesday, December 13







9:00 - 11:05 A.M.

"Parallel Computing Wi th Vertical Data"
- William Shooman, Systems Development
"The TAB SOL Concept" - T. F. Kavanagh,
General Electric Company
"Theory of Files" - Lionello Lombardi,
University of California
"Polyphase i~}erge Sorting - An Advanced
Technique" - R. L. Gilstad, MinneapolisHoneywell Regulator Co.
"The Use of Binary Computer For Data

Thursday, December 15

9:00 - 11:05 A.M.

"Data Processing Techniques in Design
Automation " - W. L. Gordon, ~innea­
2) "Impact of Automation on Digital Computer Design" - W. 4. Hannig and T.
L Mayes, General Electric Co.
3) "Calculated Waveforms For Tunnel Diode
Locked Pair" - H. R. Kaupp and D. R.
Crosby, RCA
4') "On Iterative Factorization in Network
Analysis by Digital Computer" W. H. Kim, C. V. Freiman and W. Mayeda,
Columbia University
5) "A Computer-Controlled Dynamic Servo
Test System" - V. A. Kaiser and J. L.
Whittaker, Douglas Aircraft Co.

2:00 - 4:05 P.M.

"A Computer for. Weather Data Acquisition" - Paul Meissner, J. Cunningham
and C. Kettering, National Bureau of
"A Survey of Digital Methods for Radar
Data Processing" - F. H. Krantz and W.
D. Murray, Burroughs Corp.
"The Organization and Program of the
Bj,lEWS Checkout Data Processor" A. Eugene ~iller, Auerbach Electronics
Corp. and fwiax Goldman, RCA
"Ultra-High Speed Dynamic Display System for Digital Data" - Burton G.
Tregub, ;ilelpar
''High Speed Data Transmission Systems"
- R. G. lilatteson and J. D. Barnard,
Stromberg-Carlson Company
Wednesday, December 14



2:00 - 4:30 P.M.


Thursday, December 15

2:00 - 4:05 P'.M.


"The Flying Spot Scanner as an Input
Sensor to a Character Reading System"
- J. S. Bryan, J. B. Chatten, F. P.
Keiper and C. F. Teacher, Philco Corp.
2) "Use of a Digital/Analog Arithmetic
Unit Within a Digital Compute~' Donald Wortzman, IBM
3) "PB-250 A High Speed Serial General
Purpose Digital Computer Using Magnetostrictive Delay Line Storage"
- Robert Mark Beck, Packard Bell
Computer Corp.
4) "The Instruction Unit of the Stretch
Computer" - R. T. Blosk, IBlil Poughkeepsie
5) "The Printed Motor: A New Approach
to Intermittent and Continuous ~otion
Devices in Data Processing Equipment"
- R.P. Burr. Circuit Research Company

COMPUTERS and AUTOMATION for November, 1960

Columbia Broadcasting System
New York, N.Y.
Electronic computer systems will be used
again in November 1960 as tools to help analyze returns and report up-to-the-minute probabilities on election races much faster and
more comprehensively than in any previous
Using data from the 1956 election, the
network has demonstrated how with their new
system they would have been able to make a
reasonably accurate prediction on the outcome
of the last election after receiving the first
1% of the returns -- at least an hour earlier
than they did four years ago.
Stored in an IBM 7090 computer are. the
results of a year of preliminary research by
computer mathematicians, supported by political scientists from major universities in 16
states. This mass of information covers more
than 500 precincts and 75 marginal Congressional Districts.
These voting areas were selected because
most of them report early on election night
and because they are a representative sample
of the total voting population. The information includes not only historical voting
records for every national election back to
1928 but also much other data such as racial,
religious, income and residential characteristics of the voters.
This stored information, together with
special reports that will be received directly from CBS News reporters in the same 500
early-reporting precincts on election night,
will constitute the two entirely new types of
data for the network's analyses and predict ions.
Other questions based on voter characteristics rather than past elections can also
be considered, including "How solid is the
South?"; "Is labor voting as a block?"; "Are
farm states voting as a block?"; "Is Nixon
getting more votes than the Republican House
candidates?"i "How does this compare by
regions?"; "How can the size of turnout be
explained?"; "Are voters dividing on religious lines?"; "Is Nixon winning the traditional strong Republican areas that are indispensable to him?"

not known for days or weeks. The news media
themselves have assumed the responsibiljty of
collecting unofficial returns in a few hours
after the polls close.
For this job of processing election returns, CBS News election headquarters studio
wi 11 ~e equipped with an IB1\1 RAI\IAC 305 with
many special transmitting and computing
As reports come in to the studio from
wire service teletypes and the network's own
telepnone correspondents, coded cards will
be punched on card punch machines. The cards
will be rushed by conveyor belt within eight
seconds to the 305, which ~ill search its
10 million-character disk file memory, check
the information against previous data, print
reports for the correspondents and flash the
latest national voting results on a lightboard for the video audience.

HRB-Singer, Inc.
North Plainfield, N.J.
A "weather computer" providing effective
data for efficient dealer management of fuel
delivery scheduling, in regions where fuel is
used for heating only, has been developed by
this company.
The device includes a totalizer of fuel
use, which continually indicates gallons of
fuel consumed by all customers over a selected
period of time. By comparing total gallons
consumed with the number of gallons delivered
to customers over the same period, the dealer
can tell if his deliveries are ahead or behind
his customers' needs. The machine records, in
weather units, accurate data, which is used to
schedule deliveries to customers using fuel
for heating.
Weather conditions are m~asured by a
simple-to-install, compact, roof-top device
that integrates atmospheric conditions consisting of temperature variations, sunshine,
clouds, winds, snow and rain, and relays this
data through a low-voltage lead-in wire to
the recording unit in the dealer's office.

There is no official government procedure for getting vote returns on election
night. In fact, the official results are
COMPUTERS and AUTOMATION for November, 1960



which reach full value at the end of a specified number of years or at a certain age.

Burroughs Corp.
Detroit, lrlich.

The rate book programs punch out cards
for off-line listing of the appropriate data
for selected years; an on-line listing of
all reserve and non-forfeiture information
for every possible duration of a policy is
also prepared for future reference.

The preparation of a lengthy and complex
ordinary rate book -- covering 40 insurance
plans and six riders -- has been made automatic at Minnesota Mutual Life Insurance
Company, St. Paul, Minn.
The company estimates that this project
handled by a Burroughs magnetic tape system -- represents the equivalent of 30 clerks
operating desk calculators for six months.
Savings over earlier tabulating machine and
manual methods are estimated at $35,000.
Results of the rate book program take
the form of 200 pages of numerical information, including rate book premiums and nonforfeiture values for a wide range of life
and term insurance policies, plus endowment,
retirement and annuity plans.
The Burroughs 205 automatically computes
all premiums -- standard, special class, income disability, double indemnity, etc. -for all ages of all policies. In addition,
the computer works out the cash values and
mean reserves of poliCies at various ages,
and calculates reduced paid-up policy values,
and the number of years and days a policy
will remain in effect after premium payments
have been halted.
The basic computer program which covers
life and endowment policies -- about one half
of the rate book volume -- is triggered by a
single card. This input card contains the
age limits of a policy, the description of a
particular payment plan, and tells how long
premiums are payable as well as the date on
which benefits expire.
The basic program consists of 3,000
steps, plus 50 20-word working areas which
contain various tables, interest rates, etc.
It takes the computer approximately six hours
to calculate the rates for a complete life
plan for all issue ages -- from zero to 65.
Another program handles annuities, which
includes annual premium-deferred and single
premium-immediate annuities, computed on life,
installment and cash refund bases.
Two other programs calculate tables and
information concerning income endowments and
term policies. The large assortment of term
insurance plans includes level, decreasing
term and special riders. About two hours are
required for calculation of endowment plans,

The off-line listing then becomes the
master sheet for direct offset reproduction.
Extra copies of rate book sheets are also
made so that the pertinent rate page can be
inserted into a policy, making its preparation faster, simpler.
Minnesota Mutual Life's four rate book
programs were programmed and coded by Assistant Actuary William H. Gilbert in four
A pioneer in the application of electronic data processing to insurance problems,
Minnesota Mutual Life recently completed conversion to its Consolidated Functions Approach
program -- a one-pass approach to a major
group of related applications. These include
billing and accounting for premiums and loan
interest, dividend calculation and notice
preparation, and commission calculation.
The company maintains some 220,000
policy records on magnetic tape, and uses
its Burroughs 205 to bill nearly 1,000,000
premiums a year.

Universal Controls, Inc.
New York, N. Y"o
Thousands of motion picture theatres may
soon sell admission tickets with vending machines. This is suggested by the introduction
of a "Vendaticket" machine, made by this company, which makes possible the sale of theater
tickets without the presence of an operator.
It combines the functions of an electronic
currency identifier, automatic ticket issuer,
and an electro-meChanical change maker in one
integrated mechanism.
The machine can also be adapted to sell
tickets at air terminals, bus and railroad
stations, stadiums and parks.

COMPUTERS and AUTO:MATION for November, 1960

H. M. Semarne (POOL)
Chairman, Public Relations Comm.,
Joint Users Group
Department A-270
Douglas Aircraft Co., Inc.
Santa Monica, Calif.
H. S. Bright (TUG)
Member, Public Relations Comm.,
Joint Users Group
Westinghouse Bettis Laboratories
P. O. Box 1468
Pittsburgh 30, Penna.
W. M. Carlson (AIChE)
Member, Public Relations Comm.,
Joint Users Group
Engineering Dept.
duPont Company
Wilmington 98, Del.

On Saturday, August 27, in Milwaukee, Wisc. ,
representatives of various computer user groups
. held a second meeting at which they decided that
there be formed an organization, provisionally
known as the Joint Users Group. Immediate consideration was given to an invitation to organize
within the Association for Computing Machinery.
Further action on this matter is expected at the
next Joint Users Group meeting, 16 December 1960,
in New York.
At the first meeting, held in May, the representatives of seventeen computer user groups had
agreed that the need exists for the establishment of
communication among such groups, and a resolution
defining this need had been adopted and published.
The second meeting produced almost unanimous ratification of this resolution by the fifteen
groups represented and the foLLowing statement was
"The objective of the Joint Users Group is the
establishment of communication among computer user groups to promote study, exchange
of information and cooperative effort in areas
of common interest.
"These areas include:
1. Common programming languages and
other means of communication between
computing machines.
2. Establishment and maintenance of
standards for communication and
COMPUTERS and AUTOMATION for November, 1960

distribution of computer programs
and techniques.
3. Exchange of information on problems
arising from the operation of a computer instaLLation.
4. Communication of methods and techniques for comparing the effectiveness
of computer problem solving techniques.
5. Consideration of hardware standards in
cooperation with other interested
agencies. "
At the Milwaukee meeting, an invitation was
received from the Council of the ACM to the Joint
Users Group "to organize within the ACM and to
take advantage of the facilities and services of the
ACM to implement this multiple objective".
The consensus of those present seemed to
favor some form of affiliation with the ACM. An
Affiliation Committee selected at the meeting reported the recommendation that the offer of the
ACM Council be accepted. However, in the face
of various contemplated developments, the assembly could not decide which of several possible alternative forms of affiliation would be to the advantage
of all concerned .
A committee on by-laws, a committee to
study communication between user groups, and
other working committees were appointed and
charged with reporting their recommendations at
the next meeting of the Joint Users Group. At this
meeting it is hoped the question of affiliation and
some questions of communication between groups
will be resolved. This next meeting is planned for
Friday, 16 December 1960, in New York City, one
day after the Eastern Joint Computer Conference.
For further information regarding past or
forthcoming meetings, write Jerry L. Koory,
System Development Corp., 2500 Colorado Ave. ,
Santa Monica, Calif.
Organizations represented at this meeting
ACM (Association for Computing Machinery)
ASA (American Statistical Association)
Alwac Users (Alwac III-E)
CO-OP (CDC 1604) (Note: of Control Data Corp. )
CODASYL (Conference on Data b'ystems LanguageE
CUE (Datatron 220) (Note: of Burroughs)
Datamatic-lOOO Users Organization (Note: of
Exchange (Bendix G-15)
FAST (Fieldata Systems)
LINC (Univac LARC) (Note: Univac: of Sperry Rand]

POOL (LGP-301) (Note: of Royal Precision)
POUCHE (Program Distribution Service of
the American Institute of Chemical
RCA 501 Users Group
RUG (Recomp III) (Note: of Autonetics)
SHARE (IBM 704/709/7090)
TUG (Philco 2000)
USE (Univac 1103A, 1105)
Univac Users Association

Dr. H. J. Wall
Manager of RCA Applied Research
Radio Corp. of America
Camden, N.J.
A scrambling device which makes it virtually impossible to forge passbook signatures
in the withdrawal of savings bank deposits and,
at the same time, substantially reduces customer waiting time at tellers' windows has been
Known as Signaguard, the device reproduces a passbook signature as an unrecognizable mass of broken lines. When the passbook
is presented at the teller's window, the device returns the signature to its original
appearance for comparison with the customer's
signature on the withdrawal slip.
Banks are expected to use the device in
conjunction with electronic data processing
systems situated in their main offices.
This would eliminate the need for duplicate records of signature and account status
in each branch office and also would do away
with time-consuming withdrawal checking procedures.
The new protection system makes use of
fiber optics ~- glass tubes that carry light
and images around bends and corners.
It so effectively scrambles a signature
that even a bonafide depositor, much less an
aspiring forger, cannot decipher the name on
a lost passbook, he said. The imprint is
produced as mixed segments of lines thousandths of an inch in diameter, corresponding
to the diameter of the tubes.

forgery att'empts by persons who know the depositor well and can duplicate his signature.
The new device will make savings bank
accounting foolproof. Another of its advantages is the elimination of tub files
files containing cards on all depositors -which each teller must consult before completing a withdrawal. This would make for
more room and less confusion in the banks,
and would eliminate the time required to refer to such files for signature comparison
and account checking.
Signaguard may be likened to a cable or
telephone wires, with each individual wire
being a glass tube instead of copper wire.
Each tube picks up a small segment of the
signature and transmits it to the other end
of the tube via a devious route. This
scatters signature segments throughout the
unintelligible mass that is imprinted on
sensitized paper at the other end of the tube.
In the bank the fiber optics tube is reversed to bring the scrambled signature back
to its original form for comparison with the
signature submitted to the teller.
Use of Signaguard in conjunction with
the bank's central computer system would permit a teller to complete a banking transaction without leaving his window.
The teller would check the scrambled and
withdrawal slip signatures, then punch the
necessary information into his window calculating machine, modified to feed the information via telegraph or telephone lines into
the computer in the main office. The computer would perform whatever arithmetic is
necessary and update the passbook, returning
this information directly to the teller at
the window. Should the withdrawal slip call
for more than the account balance the computer would make this known.
Under this centralized system, the depositor could transact business at any window in any branch of his bank and could
avoid a long wait because his name was Jones
and the "J" window was the only one designated
to serve him. The net result would be a reduction of the time necessary to carryon the
banking procedure by spreading the work load
among several windows.
Applications, other than in savings
banks, are contemplated for Signaguard. It
may be employed in identification cards for
commercial bank depositors, for retail credit
cards, etc.

As a double check, a secret number can be
attached to the name. This effectively foils

COMPUTERS and AUTOMATION for November, 1960



Cleveland Graphite Bronze Div.
Clevite Corp.
Cleveland, Ohio

B. H. Ciscel, General Mgr.
Chance Vought, Electronics Div.
Dallas, Texas

A new source of electricity for ignition
systems in small gasoline engines and other
purposes has been developed by this company.
The new device is smaller than a man's
fist and contains two ceramic parts that convert a single, short motion into a 20,000volt charge. In power lawn mowers, outboard
motors, portable generators, and other comr
pact power equipment, it will perform a job
that now requires a magneto, points, coil and
The Clinton Engine Corp. is introducing
the device on a new engine in its 1961 line
for power lawn and garden equipment.
The device has been named a "Spark Pump"
(trade mark) because it produces a spark each
time pressure is applied to it. A Spark Pump,
a switch, and a spark plug constitute a complete ignition system for a small motor.
In addition to simplifying the ignition
process, the Spark Pump produces a constant
high voltage at all engine speeds. This means
that many engines can be started with a single
turn of the familiar starting cord, without
the need for complex starting mechanisms.
A demonstration model of the Spark Pump,
when squeezed with ordinary handshake pressure, generates a spark hot enough to jump
across a quarter-inch gap.
The basis for the new power source is
piezoelectricity. Some natural crystals, such
as quartz, will emit a tiny electric. current
when twisted or bent. Cut into paper-thin
slices, such crystals have long been used as
the pick-up elements that translate phonograph needle vibrations into electrical impulses that can be amplified and played
through a loudspeaker.
Many years ago this company found out how
to grow these crystals artificially, and in
recent years has developed a new ceramic material (trade marked PZT) that does the same
work but better.
The Spark Pump is the first device using
piezoelectricity to create a spark twice as
powerful as that produced by an ordinary magneto and condenser.

COMPUTERS and AUTOMATION for November, 1960

A new, compact electronic device which
significantly increases the effectivenes~ of
radar for aircraft, ships or land bases has
been developed.
The device -- called a "video correlator"
makes it possible to acquire a distinct
radar target image under circumstances in
which the image otherwise would be obscured
on the radar screen, company officials said.
It achieves this by correlation of the available signal power to provide a clear target
signal at the radar's maximum operating range.
The Correlator is adaptable to sea or
land-based radar, airborne radar and other
systems. It can be installed in existing
radar systems with very little modification
and at very low cost compared to that of the
basic radar equipment.
The video correlator performs post-detector correlation of the video signal. Its
basic function is the sorting of target
pulses out of noise and interfer~nce, based
on a uniform time-spacing or delay between
successive video pulses or the pulse repetition period of the radar.
Key elements of the device are two
matched magnetostrictive delay lines and
their associated coincidence gates. All
characteristics of the raw video ~ignal are
retained without distortion.
The correlator increases radar effectiveness by sorting out and correlating only
those radar pulses which bounce back from a
target and by eliminating signals caused by
noise and interference. Since target pulses
are received at uniform time intervals and
the noise and ~nterference are either irregular or occur at a different time spacing, the
correlator, using a delay principle, selects
only those signals which come from the target.
Performance capability of the video correlator has been proved by several hundred
hours of operation both in the laboratory
and in flight tests. The device now is
available in production quantities and can
be tailored for any existing radar.


Digitronics Corp.
Albertson, Long Island, N.Y.
A new electronic system which transmits
data over the regular telephone network at a
speed of 1,500 words per minute has been developed by this company.
This new development, called the Digitronics Dial-o-verter System, was created to
function with the Bell System Data-Phone 200.

Using the new system a person can simply
dial when he is ready to send or receive data
in machine language. This system can bring
information from plants to data processing
centers, or from office to office, with more
flexibility than private telephone or telegraph lines.

The system makes possible the transmission and receipt of volumes of data, to and
from many remote locations, at high speed,
with more accuracy, and at lower cost than
has been available to date.

For greater accuracy, the system checks
the line for transmission, before it permits
data to be sent. Other features, designed
to assure reliable data transmission, include
error reporting, error retransmission options,
and double parity.

The new system can replace low-speed,
electro-mechanical equipment. Currently,
data may be transmitted over private telephone or telegraph lines at a speed of 6 to
10 characters per second, while the new system operates via Data-Phone at a speed of 150
characters per second.

The Dial-o-verter System can read or
write data via punched paper tape, punched
cards or magnetic tape. It can transmit
data in one medium at one point, and have it
received in another medium at the other

International Business Machines Corp.
New York, N.Y.
One of the largest accounting systems in
the world will soon be under automatic control, this company announced after successfully bidding for the conversion contract.
The proposed Internal Revenue Service
system involves a large computin9 center to

be established in Martinsburg, West Virginia,
with smaller computers in regional areas. It
will handle all forms of Federal ta~es and
cross check such data as individual income
tax returns, W2 forms, corporate income tax
statements, dividend payment notices, and
interest payment notices.
In preparing its bid on the system, which
will get under way in late 1961, IBM employed
a simulation program for timing and cost evaluation of a number of equipment configurations.
More than 50 systems specialists were put to
work on the voluminously detailed study.
COMPUTERS and AUTOMATION for November, 1960


Farrington Mfg. Co., developers of the first
automatic address readinn device for the
Postal Service.

The following is the "Digital Computers"
section of the program of the Annual Conference on Electrical Techniques in Medicine and
Biology held by the Institute of Radio Engineers in Washington, D. C., on October 31,
November 1-2, 1960.

At present, the machine can read only
those addresses which are printed in conventional capital and small letter style ("upper
and lower casell ) . It ignores postal zone

Session on Digital Computers
Chairman: G. N. Webb

However, new refinements are being added
which should make reading of capital letter
addresses and zone numbers possible.


Introduction to Digital Computers and
Automatic Programming, R. S. Ledley,
National Science Foundation, New York,


The use of Computers in Medical Research
at the National Institutes of Health,
N. Z. Shapiro, National Institutes of
Health, Bethesda, Md.


Problems Arising in the Digital Interpretation of Electrocardiogram Data, R. J.
Arms, National Bureau of Standards,
Washington, D. C.


Recording of Bioelectronic Signals for
Digital Computer Analysis, H. Zimmer,
Georgetown University Hospital, Washi ngton, D. C.


A Digital Computer Program for Simulating
Chemical and Biological Systems, D.
Garfinkel, University of Pennsylvania,
Philadelphia, Pa.


Computer Simulation of Reactions between
Bound Chemicals, D. Irving, P. Markstein
and J. D. Hutledge, IBM Research Center,
Yorktown Heights, N.Y.


Automatic PTogramming to Assist Simulation, J. B. Wilson, National Science
Foundation, New York, N.Y.
Post Office Department
Washington 25, D.C.

The new postal machine that "reads" the
addresses on letters is being improved to
handle addresses which are printed entirely
in capital letters -- and postal zone numbers,
too -- Postmaster General Arthur E. Summerfield
disclosed today.
The Postmaster General's disclosure followed a recent progress inspection of the machine at the Alexandria, Va. plant of the
COMPUTERS and AUTOMATION for November, 1960

This is of far-reaching significance,
Mr. Summerfield said, because a large part of
the nation's mail is addressed by mailing
equipment which prints entirely in capital
letters. This is particularly true of some
major types of business mail, such as advertising materials sent through the mails.
The necessity of reading postal zone
numbers, which are invaluable in the speedy
handling and delivery of mail, is obvious.
Including the present capacity of the
machine to read the conventional type "upper
and lower case" printed addresses, together
with its ant1cipated ability to handle capital letter addresses and zone numbers, the
device should be able to handle all kinds of
commercial correspondence quickly, the Postmaster General observed.
In addition to reading addresses on
letters, the machine sorts them into destination slots. In fact, it is able already to
read and sort 9,000 letters an hour to 40
destinations, and this capacity will be increased as the machine is developed further.
The Farrington "Automatic Address Reader"
has been under continuous development since
1954. An experimental model was successfully
tested and demonstrated by the Post Office
Department as early as 1957.
As a result, the Post Office Department
awarded a contract to the Intelligent Machines
Research Corp., Farrington subsidiary at Alexandria, for development of an even more advanced transistorized experimental model.
This is the model which the Postmaster General
viewed during his recent inspection. Delivery
of this newest device in condition to begin
preliminary testing on "live" mail is expected
early in 1961. The modifications permitting
reading of capi.tal letters and zone numbers
are expected to be incorporated by August 1961.
Through its character-sensing apparatus,
the Farrington Automatic Address Reader already is developed to the point where it can
recognize typewritten, imprinted or printed

addresses, single or double spaced, staggered
or flush in upper and lower case almost anywhere on the face of a standard size envelope.

At present the English translation emerges
in a rough but meaningful form for refinement
by human translators.

The Farrington machine "reads" an entire
city and state line, spelled out or abbreviated, as the letter passes before the "eye"
of the machine, rather than reading the address letter by letter, as required by some
commercial applications. The new development
recognizes the city and state by the general
shape of the letters, ,just as one person can
recognize another by his overall physical
appearance rather then by identifying individual features.

When more of the capacity of the photographic disk is used, it will store about
500,000 Russian words and idioms on a single
surface. Words are printed in a line code of
bars and spaces.

"Including these latest developments, I
now feel that electronic reading and sorting
of mail to fully automate postal operations,
a major objective of the Post Office Department, is just around the corner," Postmaster
General Summerfield observed.
Eastman Kodak Co.
Rochester 4, N.Y.
A high-resolution film, exposed with
present optics, is capable of storing 600
million bits of information per square inch.
This means that the entire contents of the
Encyclopedia Britannica could be stored on a
single 4-inch square piece of film.
Photography, the recording medium that
as of now stores the greatest amount of information in the smallest space, is the key
to the Air Force's new translating machine,
the Photoscopic Language Translator (see
August "Computer and Automation") that translates Russian to English at the rate of 40
words per minute.
Special glass disks, coated with a highresolution photographic emulsio~ are the
heart of the translating machine.
An entire 55,000 word vocabulary is
stored in a 3/8-inch channel printed on a
ten-inch glass disk. The channel is scanned
vertically and horizontally by electronics
until the machine matches a Russian word -fed in with punched tape -- to its English
equivalent, which is then printed automatically on a typewriter.

The glass ~isks are coated with highresolution emulsion, then inspected microscopically for any flaws in the surface.
Even with a ..?OO,OOO word vocabulary the
disk barely scratches the surface of photography's information storage capacity.
Goodyear Aircraft Corp.
Akron, Ohio
The Maritime Administration will negotiate a contract with Goodyear Aircraft Corp.
for the development of a r~dar unit that will
eliminate, or minimize, the possibility of
ship collisions at sea.
Vice Admiral Ralph E. Wilson, chairman
of the Maritime Board and Maritime Administration, U. S. Department of Commerce, said the
project is part of a 'continuing research and
development program to improve efficiency and
safety aboard American merchant vessels.
Goodyear Aircraft plans th~t the expermental radar data computer will be worked
out at its Litchfield Park, ,Ariz., installation.
The computer will be designed to plot
the projected courses of as many as 10 ships
simul taneously, 'and to sound an alarm when
collision' distances fall below predetermined
mInImums. It also would indicate an appropriate maneuver to enable the ship equipped with
the computer to avoid all vessels in the
nearby area.

Modifications of the machine will enable
it to translate more than 2,400 Russian words
per minute with greater precision and better
grammar than it does now at the slower rate.


COMPUTERS and AUTOMATION for November, 1960

For eng i neeri ng, for research, for busi ness
data processing in companies both large
and small ... The new, fully-transistorized


another outstanding computer value
... marketed by Royal MCBee Corporation

Advanced design: fully transistorized-with important
new computer design concepts that provide the largest
memory, greatest problem-solving capacity and flexibility in the low- or medium-priced field. Entire system
-computer, input-output typewriter and tape punchread console-have been specially designed as a unit.
High-speed computing ability: extra large capacity
(8008 words) magnetic drum memory, with special
fast access features. Computing speeds of up to 230,000
operations per minute. Ultra high speed input-output:
500 characters per second photoelectric punched paper
tape reader, and 300 characters per second paper tape
punch available as optional equipment. Easy to use:
maximum results can be obtained by non-technical
personnel. Users benefit from free training, continuing
assistance, an extensive library of programs. Versatile
command structure provides programming speed and
flexibility. Low in cost: priced just above the smallscale computers, the RPC-4000 outperforms computers
costing many times more. Economical to install and
COMPUTERS alld AUTOMATION for November, 1960

operate: no site preparation, air-conditioning or special

maintenance required. Plugs into any standard wall
outlet. Multiple application ability: designed to perform engineering, scientific and research calculations,
as well as business data processing and management
control functions.
The RPC-4000 is a product of the Royal Precision
Corporation, and is marketed by the Data Processing
Division of Royal McBee. It is the latest member of the
growing family of electronic computers from the people whose LGP-30 has become the world's leading
small-scale computer.

Royal Precision Corporation

Royal Precision is jointly owned by the Royal McBee and General Precision Equipment Corporations. RPC-4000 sales and service are available coast-to-coast, in Canada and abroad through
Royal McBee Data Processing offices. For full specifications, write
ROYAL MCBEE CORPORATION, data processing division, Port Chester, N.Y.


Undetected Errors ln
5-Unit Code

Transmission and Their Elimination
James F. Holmes
Lybrand, Ross, and Montgomery
New York 4, N.Y.

A great many articles are being published concerning
data communications as input-output media to data processing operations. The majority of them cover high-speed
communications using microwave and 6, 7, and 8 unit
codes with speeds in excess of 3000 bits per second. What
about the old work horse, the five-unit code for telegraphic
communications up to 100 words per minute?

The five-unit, or Baudot, code has been in use for a long
time. Telegraphic communications serve a definite need,
especially for those who require continual communications
for general as well as for data purposes and whose operations
are scattered and individually not large enough to support
high-speed equipment.
Admittedly, the Baudot code causes some problems,
especially in detecting errors. There is, however, a solution and, wonder of wonders in this day and age, the costs
of the solution are either negligible or zero. The following
discussion deals with the solution to the "undetected" error
The tests referred to were run by Western Union Telegraph Co. American Telephone and Telegraph has stated
that they will participate in a code reassignment discussion
if enough interest can be generated. Work in the area of
code reassignment is now being done in Europe on the
basis of a complete change. This paper proposes a code
called the "L" code which appears to be the simplest and
to require the least number of changes. A system will be
installed in late 1960 in the United States using the "L"
The 5-Unit Baudot Code
The punched paper tape 5-unit Baudot code has been in
use in telegraphic communications for approximately 80
years, and has proved reliable for administrative message
traffic. However, data transmission requirements in recent
years have shown certain inadequacies existing with the
standard Baudot code assignments.
The original assignment of codes was made on the basis
that the most commonly used letters should be assigned
the simplest codes. This assignment leads to some rather
peculiar and interesting transmission errors. Table 1 shows
a standard assignment of coding.

How will a character be changed upon the loss or gain
of a single pulse? It is possible to make a schedule showing
just which characters will result from a given character by
the gain or loss of a single pulse.
Control codes for data transmission must be chosen very
carefully, as will be noted from such a schedule.
For example, a particular data application demonstrating
unfortunate choice of control codes is the choice of M for
one function and N to denote an opposite function. For
instance, "M" may be used to denote manual input requiring special handling or card insertion at the data center,
while "N" may denote normal input for fully automatic
processing. The code for M is - - 0 0 0; the code
for N is - - 0 0 -; accordingly, an "M" or "N" may
be changed each to the other through the loss or gain of a
single pulse (one to the other) in the standard code assignment.
More complicating errors are possible with the assignment of upper case (numeric characters). The loss or gain
of a single pulse will convert




8 to 7 or 0
9 to 5
7 to 1 or 8

o to

1 or 8

2 to 1
6 to 1
1 to 0, 6, 2 or 7

This is a vast array of possible numeric to numeric changes
which are difficult to detect under the most ideal conditions
and may in themselves produce compensating ~rrors. The
possibility of these simple changes which may be compensating has raised the demand for error-detectihg devices
from every quarter.
Error Detection and Correction
The equipment manufacturers are willing and eager to
produce error-detecting and error-correcting devices. Many
designs are presently available and some have been marketed. However, basic economics indicate that there will be
no mad rush to install such devices. Inasmuch as the
COMPUTERS and AUTOMATION for November, 1960

Table 1
(In the codes as shown, a "0" stands for a punched hole
or a pulse and a "-" stands for no punch or no pulse.)
5 Singles:

10 Doubles:






















o o
0 o



o 0

o o



1 Blank:



























o o 0
o 0
o o
1 Quintuple:








o 0













o o








5 Quadruples:


o 0


10 Triples:





Errors fall into two general categories, i.e., detected
errors, and undetected errors. In communications terminology generally a detected error is one in which a character
is changed to a completely invalid or non-sense character;
and an undetected error is one in which a character is
changed to another valid character such as numeric converted to numeric. A relatively high detected-error rate may
be tolerable while an undetected-error rate of any magnitude is intolerable. If it is assumed that normal detectable
transmission error rates are tolerable in present accounting
applications, it then is necessary to eliminate the relatively
small proportion of undetected errors.
Most available error-detecting devices require manual
intervention and are generally inefficient in "line time."
The system may be one in which transmission continues
until an "end of block" signal indicates a check point, at
which time an error may be detected, stops both the transmitter and the reperforator and, at the same time, actuates
a visual and audible signaL The signal calls an operator,
who manually resets the received tape to the end of the
previous block, "letters out" the errored block and signals
for a re-send of the errored block. This type of operation
may prove satisfactory in point-to-point transmission where
high utilization of equipment and circuitry is not a requirement but, in any case, it is an inefficient use of personnel,
equipment and circuits and is not practicable for "switched"
tape systems.
Many systems, even those large enough for switching
operations, cannot afford the personnel time, equipment or
circuits necessary to ensure "error free" transmissions. Frequently, private wire telegraph systems are installed for
transmission of both administrative and data messages, on
the theory that the data will "ride free" on previous administrative communications costs. This means high utilization of circuits, personnel, and equipment.
Requirements for Ideal Error Detection


The ideal error detection system for data messages on a
5-unit telegraph system, whether it be TWX, WUX, Telex
or private wire, will have the following characteristics:

W ill cost nothing


Will not add to line time


Will eliminate compensating errors


W ill indicate all errors


Will provide for message reconstruction at
the receive end.


number of errors generated in the communications system
is relatively small, some of the economic factors to consider

The human error rate in data creation is the biggest variable.


Communications-equipment error rate is lower
than that in the conventional data-conversion
equipment, such as tape-to-card converters.


The information transmitted is, in almost every
case, accounting information made up of a vast
number of individual problems. Error rates are
tolerated on a conventional tabulating processing
system; and the communications system does not
normally introduce any appreciable change in the
tolerable error rate.

COMPUTERS and AUTOMATION for November, 1960

Results of Tests
Recent studies have indicated that an error detection
system is possible which will closely follow the above five
requirements. These studies have shown that:

Ninety percent of all errors are the result of the
loss or gain of a single pulse with about 88 percent being the loss of a single pulse;


Individual circuits exhibit a general pattern; for
example, an error on a particular circuit may generally be the loss of the fifth pulse;

3. Error rates for data messages on a telegraph communications system can be as low as 1 in 500,000

Undetected errors are about 1 in 7 of all errors.1

The economic problems involved in providing low-cost
error detecting devices, coupled with the generally low
incidence of error in normal telegraphic communications,
indicates that an approach other than additional equipment
is needed to overcome the undetected error problem. A
recent study has shown that a system does exist within the
structure of the 5-unit code to reduce the undetected error
rate to a maximum of 1 in 500. 1
Reassignment of Coding
In the Baudot code (Table 1), the fact that almost all
errors are the result of the loss or gain of a single pulse,
coupled with the 5-channel code configuration, provides an
economical solution to the undetected error problem. The
solution is a reassignment of coding to place all numeric
characters into the 10 triples. The keyboard layout remains
the same.
The coding reassignment requiring the least mechanical
change is an exchange of six pairs of characters listed below:


and R
Band 0

C and I
E and F

G andQ
M and T

A calculation shows that this code reassignment would
prevent the change of a numeric character to another numeric character through the loss or gain of a single pulse.
Further analysis, however, shows that an exchange of vowel
for vowel has also been eliminated, with the exception of a
possible change of an A to U and vice versa. In order to
keep the cost of mechanical coding changes to existing
equipment at a minimum, it may be assumed that the elimination of the A to U possibility may be disregarded.
In order to follow through with the reduction of undetected alphabetic errors, it would be desirable to eliminate
such errors as changing full to fall, fur to far, Jun. to Jan.;
so it is reasonable to propose one additional subsequent
coding change, A and F. A calculation then shows no change
of vowel to vowel or numeric to numeric for the loss or gain
of any single pulse. In other words, when an error is encountered, the proposed coding change will yield a consonant for a vowel and vice versa or, in the case of numerics,
a non-sense or invalid character in place of the numeric, thus
virtually eliminating compensating numerical error possibilities. Numerical errors may be easily detected in existing
tape-to-card and tape-to-magnetic-tape converters, computers and other devices, using the proposed reassigned coding.
The L Code
The actual resulting changes in the A code, producing
the L code, are shown in Table 2.
The effect of the L code can be shown in an illustrative
example as follows, assuming an extremely poor system
with an error rate of 1 in 10,000 characters, consisting of
14 circuits operating at 75 words per minute, 9 hours per
day at 90% utilization:


The L Code

The A Code

100% transmission 3,402,000 char/day 3,402,000 char/day
90% utilization

3,061,800 char/day 3,061,800 char/day

Errored characters

306 char/day

306 char/day

Undetected errors

44 char/day

0* char/day

* (An undetected error might be expected once every 11 days
on the basis of a maximum rate of 1 in SOO.)

Table 2





















The proposed reassigned coding (referred to hereinafter
as "L" - for "logical" - code) is essentially a data-transmission requirement. It is not to be expected that existing
5-channel systems would be changed to the "L" code immediately or in the foreseeable future but, rather, that "off
line" data conversion equipment be changed. There is,
however, no reason not to install entire new systems which
would use the "L" code.
Equipment is presently available which can be used "on
line" for transmission in the standard code for administrative purposes and includes the use of a third shift to provide numeric data transmission in a reassigned code. 2
Loss of Complete Characters
There is an additional area of error in data transmission
that is encountered in normal operation. This is the loss
of a complete character, shifting data out of field without
actual loss of any significant digit by such happenings as
the dropping of spaces affecting the field which defines the
numerical unit. For example, a 1 shifted a column to the
left will be interpreted by computers or data equipment
as 10.
A simple pulse generator connected to the receiving leg
will keep the receiving equipment in synchronization with
the previous rhythm before loss of line signaL This will
maintain synchronization upon restoration of the line signal, and will insert blanks for the lost characters in the received tape. The blanks can then be detected as errors
using other equipment.
COMPUTERS and AUTOMATION for November, 1960

The cost of the pulse generator or teleprinter synchronizing set 3 will be proportional to the number of start-stop
pulses that must be generated upon signal loss. Equipment
is presently available that will generate up to 20 startstop pulses a but for normal day-to-day operation synchronizing equipment producing 3 to 5 start-stop pulses
would be sufficient. Radio transmissions would require the
higher-order teleprinter synchronizing set.
A small inexpensive pulse generator could be incorporated within the receiving unit distributor circuitry at a
very minor increase in monthly rental.
The teleprinter equipment most readily adaptable to the
"L" coding is the type 28 line of teleprinter equipment
manufactured by the Teletype Corporation. Studies are
presently under way to determine the COSt of conversion of
existing type 28 equipment.
Other off-line data-preparation equipment may be obtained from the manufacturer using the "L" coding at no
increase in costs. Tape-to-card converters can be made
adaptable to the "L" code through the plugboard.

The "L" code will provide the following:

Rapid error detection


Possibility of message (character) reconstruction
at the receiving end

rattling good history

3. Available at little or no COSt
4. Can be transmitted on existing systems utilizing
off-line equipment for data preparation

5. Will eliminate undetected errors

6. Will eliminate compensating errors

Does not increase line time.

The proposed adoption of the "L" code does not provide
for on-line error correction, which is felt to be time consuming and uneconomical. The inclusion of teleprinter
synchronizing circuitry with the receiving distributor will
maintain synchronization upon loss of the line signal.
The adoption of the "L" code and inclusion of a simple
teleprinter synchronization device would virtually eliminate
the problems existing in transmission of data using standard
teleprinter circuits and equipment.

Error-Checking Possibilities Concealed Within the 5-unit Code,
by R. Steeneek, Western Union Technical Review, Vol. 14, No.
2, April, 1960, pp. 69-71.


Teleprinter for Reliable Transmission of Numbers, by S. Augustin, Electrical Communications, Vol. 35, No.4, 1959, pp.


Teleprinter Synchronizing Set SYZ-634, by W. Schiebeler,
Electrical Communications, Vol. 35, No.4, 1959, pp. 247-250.

The author acknowledges with gratitude the assistance
of The Western Union Telegraph Co. in obtaining test data.
COMPUTERS and AUTOMATION for November, 1960

"War," wrote Thomas Hardy, Hmakes
rattling good history; but Peace is poor
reading." Scientists at Pro}ect Omega,
in Washington, taking a pioneering part
in the ancient and honorable tradition of
war gaming that stretches from the first
chess of 3,000 years ago to modern
stochastic models, are writing rattling
good history in both fields,
war and peace.
Synthetic history, they call it: the
application of advanced mathematical
thought, and the digital computer
simulation of war or in support of map
battles, have brought Pro}ect
Omega to the frontiers of new
developments in gaming, for Army,
Navy, Air Force, OCDM and ARPA, as
well as business and industrial sponsors.
For your free copy of THE GAME OF
WAR, an illustrated history of the
highlights of war gaming over 3,000
years, illustrated with authentic warriors
of the periods, write to James L. Jenkins.

Technical Operations,


3600 M Street Northwest Washington 7, D. C.


Part 1
James T. Culbertson, Math. Dept.
California State Polytechnic College
San Louis Obispo, Calif.
The term "automation" is a new word introduced about
1947 to replace the more unwieldy word "automatization."
The limits to its meaning are not yet settled, but roughly
speaking, any improvement in process control by automatic
devices constitutes automation.
An early clear-cut example of automation is the discovery
in 1713 of automatic valve control by the ingenious boy,
Humphry Potter, who had the monotonous job of opening
and closing the valves on a Newcomen steam engine. He
found a way to fasten the valve handles to the piston rod
by wires so that the valves were turned on and off at the
right moments, thus replacing human control by automatic
cOntrol. Later automation of the steam engine included
automatic control of its speed by Watt's governor, and
modern examples are roo numerous and well known to be
worth mention.
Ordinarily "automation" is any improvement in the
control of some activity or process by non-human, i.e., automatic means, but sometimes the term is defined more narrowly. Thus some say there is no automation unless the
control is based on feedback or self-correction. This would
exclude many cases of automatic control, e.g., where the
steps in some process are timed and sequence-controlled by,
say, magnetic tape (there being no feedback except perhaps
in some trivial sense); this happens in various completely
automatic milling operations. Feedback is usually involved,
however, both in complicated and simple cases of automatic
control. Information about the output is continuously fed
back into the servomechanism which corrects its output
Another limitation on the use of the word "automation"
is that sometimes it means merely "the use of electronic
equipment," but this usage imposes an undesirable restriction. The precise definition is of not toO much importance,
however, since what actually constitutes automation is
fairly well agreed upon. But we should be careful to distinguish between automation and mechanization, which
designates a more primative process. Likewise, mechanization is to be distinguished from the even more primitive
tool using.
Starting at the very beginning, we may think of ten
stages leading to complete automation. These are rather
arbitrarily chosen and overlap considerably, but they may
help give a picture of the developments leading up to the
present and into the future.
Stage 1. No Agents.
Life on the earth is believed to have started very gradually developing from the "primordial soup" about 1500

million years ago. Prior to that time there were no individuals of any kind, i.e., no agents. There was nobody doing
Stage 2. Agents but No Tools.
First complex self-perpetuating chemicals, then more and
more complex viruses, bacteria, and single celled organisms
arose, and there were, definitely, individuals or agents that
were active or doing things. A very gradual evolution proceeded for many millions of years and animals became more
and more complex. There were individuals or agents but
they used no implements. There was just behavior, bodies
moving around, but no artificial bodily extensions, or tools.
We may be safe in believing that prior to, say, 200 million
B. C. no tool was ever used except perhaps accidentally or
in some trivial way.
Stage 3. Agents Occasionally' Using Tools but
Never Making Tools.
As animals evolved further, their behavior became more
complex and, at this stage, natural objects conveniently at
hand were occasionally used as tools.
One of the most interesting present-day examples is the
burrowing wasp, Ammophila. The female digs out a burrow, lays her eggs in it, then stings a caterpillar which she
also puts in as food for the young when they hatch. She
then seals the entrance with dirt, and finishes off the job
by hammering down the dirt with a pebble. Here a suitable pebble is selected and definitely used as a tool.
Except for a very few cases like this, however, the use of
implements by invertebrates, and in fact by any animals
below the birds and mammals, seems practically nonexistent.
The Darwin finches of the Galapagos Islands are persistent tool users. These birds feed on insects embedded in
the bark of trees. They select cactus spines which they
hold in their beaks and use to pry out their victims. If
spines cannot be found, they will use other handy utensils
such as twigs, which they can break off from bushes. 1
Elephants sometimes break off branches with which to
scratch their backs. Monkeys like to roll stones down hill
on those who take too great an interest' in them. Baboons
throw stones at scorpions to kill them. Apes sometimes use
a straw as a dining utensil, putting the end of the straw on
an ant hill, waiting until a line of ants walks onto it, and
then licking off the ants 2. Also they use sticks to get outof-reach objects.
"With my gun-bearer N'Gombie I crept through the
bush and found eight chimpanzees - six of them alCOMPUTERS and AUTOMATION for November, 1960

most full-grown - sitting in a circle at the edge of a
small clearing. . . . They were making a lot of noise
and kept beating and pushing each other aside, but for a
time I could not see what they were up to. . . . Watching through my binoculars, I could see that the chimps
were sitting round the entrance to one of these nests.
Each ape held a long twig, which it poked down the hole
and withdrew coated with honey. There was only one
hole, and though for the most parr they took turns at
using their twigs, quarrels were constantly breaking out,
and those who had licked off most of their honey tried
to snatch the newly-coated twigs. We watched them for
over half an hour at a range of fifty yards, before creeping away as silently as we had come, so as not to disturb
the party. This is one of the few examples I have known
of a wild animal employing a tool." 3
We need not consider, of course, the cases where domestic or captive animals are taught to use implements.
It is only with the evolution of man that we find any
extensive use of implements. Anthropologists believe that
man's ancestors became quite proficient users of ready-tohand weapons and other implements at least ten million
years before they made any of them.
"There is evidence suggesting that some of the early
Hominidae beginning to walk upright on open ground but
possessing brains no larger than typical apes, may have
been anatomically well enough equipped to use tools. Insofar as they were adapted to life on the ground, they
would be capable of using improvised tools and weapons,
as do baboons and chimpanzees when circumstances demand.
. . . The Hominidae may have remained for millions of
years at the stage of occasional tool-users." 1
Stage 4. Agents Collecting and Using Tools but
not Making Tools.
Using implements more and more frequently man's
ancestors reached a point where they became collectors of
sharp stones for cutting food, nice sized stones for throwing, handy sticks for clubs, and so forth. There does not
seem to be any record of any tool collecting by the lower
animals, except perhaps for stones in the crops of poultry
and a few similar borderline cases.
Stage 5. Agents Making Muscle-Powered Tools.
Except for a very few extremely simple examples, there
seems to be no tool making by animals. Psychologists have
shown that chimpanzees sometimes have the insight to
assemble a very simple implement. In one experiment twO
sticks are in the cage and the ape tries first one of them
and then the other but neither is long enough to reach the
bananas outside the cage. But one stick is hollowed out at
one end so the other stick can fit into it to make one long
stick. The ape discovers this and then uses the long stick
to get the bananas. After doing it once, of course he does
it very quickly next time, and will even chew a stick down
a bit if it doesn't fit into the other one 2. This is about the
limit, though, of "tool making" among the apes, and animals below the apes do not seem to make any tools at all.
Man is practically unique as a tool-making animal.
It has been estimated that human tool-making began
about two million years ago. At first it was irregular and
incidental but then about one million B.C it became a
routine daily procedure or tribal occupation.
COMPUTERS and AUTOMATION for November, 1960

Stone Age. The first artifacts were stone, and stone
remained the favorite raw material for a remarkably long
time. The stone age ended at different times in different
places. Thus the stone age lasted until 3000 B.C with the
Greeks and until 2000 B.C in Western Europe, while the
Australian natives were still in the stone age one hundred
years ago.
The first standard tool of the Old Stone Age was a large
almond-shaped general-purpose implement chipped so as
to fit nicely into the hand. From .this primitive hand ax
other more specialized implements such as scrapers, cutters, borers, and shaped and pointed flints, and the mortar
and pestle gradually evolved.
Although men made only stone instruments at first,
they soon used these primary tools to work other material,
especially wood, into implements long before the stone age
ended. Putting a wooden handle on the ax was a tremendous step forward, and then came flint-tipped spears
and other utensils and weapons. Progress was extremely
slow by modern standards.
In the New Stone Age (10,000 to 2,000 B.C in Western
Europe) the stone implements became polished and specialized, and also there were wooden mallets, wedges,
needles, digging sticks, fire drills, paddles, ladles, shovels,
boomerangs, etc. During this period the bow and arrow and
the wheel were invented and hunters made rather ingenious
traps and animal snares.
Bronze Age. Copper was the first metal to be used as
a raw material. The discovery that it could be melted and
poured into stone or clay molds revolutionized the toolmaking industry. Tin was soon added to form bronze,
which was harder and made better edged tools and weapons.
The first metal implements were made in the Near East
approximately 4000 B.C; for the Greeks the Bronze Age
began about 3000 B.C Implements of the Stone Age were
re-made with bronze, and in addition there were made the
first swords, picks, files, saws, oars to row with (instead of
just paddles), plows, looms for weaving, potters' wheels,
and a great variety of new utensils such as hair brushes,
paint brushes, balances for weighing, calipers, and writing
and drawing instruments. The first cities were built and
man began to lead a more complicated life.
The first devices using animal power were perhaps the
New Stone Age sledges, but later plows and war chariots
were also animal powered.
Iron Age. About 1000 B.C man learned to smelt iron
ores, a very abundant raw material, and the number and
variety of implements was greatly increased. For the first
time there were hinged tongs, scissors, pliers, anvils for
making nails, metal drills, blocks for drawing wire, planes,
simple and compound pulleys, cog wheels, water screws
and bucket wheels for irrigation, presses for extracting olive
oil, battering rams, etc. Improvements were made in all
types of implements.
Medieval and Modern Times. Medieval to modern times
added to the muscle-powered tools and implements: improved hand presses (for printing), hand pumps ( to
remove water from mines), tread mills, hydraulic presses,
a new type of lock and key, electrostatic machines, the
obstetrical forceps and many other instruments.
All tools under discussion here are muscle-powered,
though many power-driven devices, discussed in stage 6,
had already been made by this time.

At an IBM Datacenter, customer programmers use ready-made programs to help solve
data processing problems.

The Programmed Applications Library will make
complete industry pro-

grams available.


'ghost writers' for computers
Expert IBM programmers help customers cut costs
Working behind the scenes at IBM is an unusual group
of over 300 men and women. They're specialists who
develop computer instructions, or "programs."
It is what they "write" that lets you translate
business and engineering language into the language that makes sense to a computer.
Programming is a demanding process. And it
takes time. It took, for example, the equivalent of
45 man-years to develop the FORTRAN (FORmula
TRANslator) systems which IBM released in 1957.
Containing 75,000 programmed instructions, FORTRAN is still the most widely used system for solving scientific problems on IBM computers.
New system to use everyday business language
IBM programmers are now implementing COBOL-x-,
the COmmon Business Oriented Language translator. This, like the IBM Commercial Translator introduced last year, is a programming system in
which everyday business language becomes the language of computer instructions. The result of all
these machine languages is a saving in programming costs that might easily equal 40% of a customer's data processing system investment.

Other methods of lowering your costs

vious advantages, use of these systems permits you
to put your computer to work sooner ... and frees your
staff to concentrate on systems design and methods
of operation, instead of on programming details.
IBM offers the largest free selection of programming systems available to industry. The list is being
enlarged and the programs improved almost daily.

You can get 2,000 tested programs
Many IBM customers are members of SHARE and
GUIDE ... progressive and independent IBM user
organizations. Through these groups, they exchange experiences and ideas, and eliminate duplication of programming effort. SHARE and GUIDE
members developed and contributed a large number of the 2,500 tested programs currently available
through the IBM Program Library. This library is
established to distribute IBM-written programs
and programs contributed by IBM customers.
Your IBM Representative will help you put these
programs to work. Ask him about them.
*This is a joint effort of computer
manufacturers and users under
the guidance of the Department
of Defense.

Every IBM customer is entitled to many programming systems on a no-cost basis. In addition to the ob-

balanced data processing

, pro-


Programming routines that translate business
problems into computer language are provided
to IBM customers without charge.

An IBM programmer
teaches customers the latest programming techniques.

The construction of muscle-powered devices continues
to the present with typewriters, hand-cranked ditto machines, hand punches, bumper jacks, rotary can openers,
and various other gadgets.
Many muscle-powered machines of not long ago, such
as threshing machines, horse-drawn reapers and binders,
pedal-driven sewing machines and lathes, hand-operated
wringers, old fashioned hair clippers, and cranked adding
machines have now become almost obsolete because of
The humanly operated machines are extensions of our
limbs and fingers such that our muscular power is increased
or directed in some improved way. Thus the club or lever
may be thought of as an extension of the arm. Likewise
the needle is better than mere fingers in pushing thread
through cloth, and we can mix eggs better using a hand
egg beater than we can using just our fingers.
Stage 6. Mechanization. Driving the Tools with
Non-muscular Power.
Mechanization means the use of water power, steam,
electricity, or some other non-muscular force to drive tools,
implements or machinery of any kind.
In some cases the power may be stored up indirectly
by muscle power, as when a watch is wound by hand. Thereafter, however, it runs directly by tension in the spring and
is considered an example of mechanization; similarly, for
pendulum clocks wound up by lifting weights. In a previous article we discussed the history of these watches, clocks,
orreries and automata. 4 They are relatively weak though
powered cousins to the power-driven machinery of the industrial revolution.
The great age of mechanization began about 1750 in
England with the first Industrial Revolution or widespread
transition from muscle-powered machines to power-driven
machinery, powered by water, steam and later electricity.
But examples of mechanization extend back far before
this. Earliest perhaps, but quite trivial, is gravity, such as
rolling wagons down a hill and drifting boats down a
stream. Later came a very simple but commercially important example - the discovery or invention of the sail
boat during the Bronze Age.
The first full-fledged case of mechanization seems to
be the water mill introduced in the Iron Age, principally
for grinding grain into flour. The first primitive water
wheels developed 3 or 4 hp., but the Roman engineers soon
made wheels yielding up to 60 hp. The miller living by
the stream became an important industrialist. A little later
windmills were also used. Muscle, wind, and water however remained the only industrial prime movers until steam
power was introduced. One other vital, or rather lethal,
example prior to the Industrial Revolution needs to be
mentioned, namely, the gun, which works on the same
principle as the internal combustion engine except that
instead of the piston there is a bullet. Firearms were invented about 1330.
Large factory industrialization began with some inventions in the cotton textile industry - the flying shuttle
(1733), spinning jenny (1765), power loom (1780) and
so forth. Since then there has been a steady increase
in mechanization in all fields. Food processing plants,
chemical plants, glass and bottling works, ceramic, plastics
and tool-making factories, quarries, foundries, tanneries,

breweries, wineries, distilleries, dairies, prmtmg plants,
bakeries, canneries, sugar refineries, laundries, silk mills,
steel mills, saw mills, paper mills, and many others became partly mechanized and then more and more mechanized. Large factories were built. For practically all industrial tools, power other than muscular could be used
as the economics of the situation demanded. These developments are well known and too numerous to be worth
mentioning. This mechanization was greatly enhanced by the
two accompanying industrial improvements - interchangeable parts and the assembly line. Big industries arose in
England, Europe and America.
The first uses of steam power for mechanization seem to
be Somerset's "water commanding engine" for lifting water
( 1650) and Savory's steam pump for mines (1698). Newcomen's steam engine (about 1710) made 12 strokes per
minute and developed about 6 hp. Later Newcomen engines
had 15 inch cylinders and were rated up to 80 hp. Watt
invented the modern piston and cylinder steam engine
about 1760, making the first really successful one in 1780,
which led to a practical locomotive in 1830 and then the
steamboat. Several decades later big high-speed steam and
hydraulic turbines led to large generators for the new, much
handier, prime mover, electricity. Though the dynamo was
invented by Faraday in 1831, there were no commercial
power plants until the 1880's. First used for Edison's
electric lights, these rapidly provided more and more power
for industry. Before the turn of the century came gasoline
and internal combustion. Now we have another very important substitute for muscular power, nuclear energy, extremely powerful and perhaps the last prime mover.
To call such prime movers "substitutes" for muscle is,
of course, an understatement. They also extrapolate the
muscular possibilities. No amount of muscle could throw
a satellite into an orbit.
Stage 7. Simple Automation. Artificial Guidance
or Control of Processes, Tools, or Machinery.
Automation implies replacing human observation and
guidance by artificial observation and guidance. Usually
automation means automatic control of power driven machinery - machines running other machines by automatic
controls. Thus most cases of automation are also cases of
mechanization. As is often said, mechanization replaces
human brawn while automation goes a step further and
replaces human brain. Mechanization and automation together can replace men by machines in industry.
Borderline cases are hard to classify. Are Hero's lamps
(60 A.D.) with automatically advancing wicks examples
of automation? Hero also made a heat-operated device
which opened temple doors after the priest had lit the
fire and closed them when the fire died out.
The Roman mileage indicator is a good example of
automation. At first road distances were recorded by
counting wheel-revolutions, one spoke of a chariot wheel
being painted black. This was tedious and also difficult,
especially with the Roman number system and a fast driver.
Then about 100 A.D. they invented a famous automatic
mileage indicator, so devised that one marble dropped
through a hole with each revolution of the wheel. Counting
marbles at the end of the trip gave the distance traveled.
[To be continued in the December issue of
Computers and Automation]
COMPUTERS and AUTOMATION for November, 1960

Do you get
what you pay for
with Honeywell 800 ...?

The ability to do up to
8 J·obs simultaneously

We say Honeywell 800 will process more data per dollar per working day than any high-speed electronic system
now on any office floor (or on any drawing board). We can afford to say this, because Honeywell 800 lets you process
up to 8 jobs all at once. And we're including big, company-size jobs - like running off a payroll, scheduling production, controlling inventory, or preparing a complete sales analysis. Each job has its own program, too. No complicated
system of priorities or costly super-program is needed. In a way, you get up to 8 individual computers to work with.
You can see how this versatility (we call it Automatic Parallel
Processing) cuts operating time and costs to the bone. To find out
more about this Automatic Parallel Processing and many other
design cc firsts" built into Honeywell 800, just write MinneapolisHoneywell, Datamatic Division, Wellesley Hills 81, Mass.

II e~D~p~~

COMPUTERS and AUTOMATION for November, 1960


d· ers'and,
Editor s Foru

The front cover shows a "digital computer demonstrator,"
in a lucite case, for use in schools, lecturing, etc. It is able to
add, subtract, multiply, and divide in the binary number
system. It does not have a memory; it has a single input.
The theory and procedure are explained in a manual. Total
weight, 12 pounds; power consumption, about 5 watts; size,
19" by 19" by 13". It is made by Aironics, Inc., Hialeah,



for 10- 12 , related to the printer's term pica, meaning "a point," and probably related
to the French pic, a point


for 10- 9 , from the Greek nanos, "a dwarf"


for 10 9 , from the Latin gigas, "a giant"


for 10 12 , from the Greek teras, "a monster"



S. F. Grisoff
Poughkeepsie, N.Y.
With reference to the article on page 8B of the August 2
issue, "Daedalus: Complete-the-Square Computer" I wish
to point Out an erroneous statement made.
The last statement says that "a skilled human player can
outplay the machine - especially if he has the first move."
This is not so. In this game with four dots on a side, it
can be proved that the second player has a strategy which
will assure him getting the majority of points (five in this
case). Furthermore, the proof can be extended to the game
with n dots on a side (n > 2), and in this case the second
player is assured either of [( n - 1) 2 + 1J /2 points (a
majority) if n is even, or of (n - 1) 2/2 points (exactly
half) if n is odd. Note that these are his minimum scores
using his optimal strategy, and if the first player does not
play exactly right, the second player may get many more.
The optimal strategy mentioned is simply that the second
player moves circularly symmetric to the first player's last
move. The results of this are obvious.
The optimal strategy for the first player is to sacrifice a
box immediately putting the burden of play on the second
player. However, he can never make up that box. If he
does not do this he will lose more than he should.

o. Hugo Schuck
Director of Research
Minneapolis Honeywell Regulator Group
Minneapolis 40, Minn.
I. From

I am greatly impressed with the picture on the front cover
of your September, 1960, issue of Computers and Automation. It is the best I have seen of the plenary session of
the IFAC Congress, which I attended as a U.S. delegate.
The picture I took is not nearly so good, nor wide angle,
and I would much like to get a copy of yours. Will you
please advise what you can do to make my record more
complete, as I use it to give talks to various interested community groups?
II. From The Editor
I am glad that you liked the picture on the front cover of
the September 1960 issue of Computers and Automation.
This picture came to us from the Press Department of the
Soviet Embassy, Washington, D.C.; as a matter of fact it was
assembled from two pictures. I would suggest that you
ask them for prints of the twO photographs from which our
front cover was made, and I feel sure that they will cooperate with you and give you prints.
(Part 2)

Frank Leary
Associate Editor, Electronics
McGraw-Hill Publishing Co.
New York, N.Y.
In connection with your glossary of terms, the scientific
community, urged by the National Bureau of Standards, is
adopting four new prefixes:

(Continued from the October 1960 issue of Computers
and Automation, page 32)
Edmund C. Berkeley
(Condensed Report)
The president of a large eastern college said the other
day to a friend of mine, a professor there, "I think your
COMPUTERS and AUTOMATION for November, 1960

sugge~Lion of having a Peace Engineering Group of 30 or

40 people here at the university is excellent. I should like
for it to be started. Weare doing so much work here connected with defense purposes and with government purposes in general that I think a Peace Engineering Group
would be desirable, so as to balance the situation."
This incident struck me as a very good indication of what
is happening in this country. I think the argument about
the necessity for peace is just about won. The deterrent of
modern weapons has become so great that no rational
country dares to resort to war. The United Nations is increasingly encouraged to intervene, work, and conciliate so
as to avoid war. Many people in the U.S. government and
many people all through the United States, I believe, have
concluded that peace is necessary, and believe that it is
possible to achieve peace, and to remove the threat of nuclear death for everybody in the world, via a process of controlled disarmament with inspection.
The change in the underlying opinion of the country as reflected by that college president - is particularly important because it contradicts a good deal of what passes
for news in the headlines of newspapers.
With regard to the duties, responsibilities, and professional capacities of people in the computer field, we are
arriving at the new area of application: controlled disarmament with inspection; conversion from war industry to
peace industry. This greatly enlarges the territory in which
computers can be applied.
In the July issue of "Computers and Automation" appeared an article "Computers and Data Processing in a
National Peace Agency." The bulk of this article was verbatim reproduction of a bill, HR9305, calling for a National
Peace Agency, and indicates a number of new areas for the
application of computers and data processing.
One of the specific provisions: "The Agency shall undertake programs to carry out the purpose of this act, including among others, programs . . . for development and application of communications and advanced computer techniques for analyzing the problems involved in inspection
of national budgets and economic indicators as they bear
upon disarmament inspection systems."
Computer people may be expected to see the importance
of devoting some of their time, energy, and professional
capacity to the applications of computers and data processing in peace engineering. This is a good objective for computer people eager to make contributions to the social
benefits of computers and data processors.
Because of interest at the end of the first meeting, it was
decided to hold a "hall of discussion" on the following evening. Among the topics discussed or scheduled for discussion that evening were the following:
1. Is it possible to define a code of ethics applicable to
the social responsibilities of computer people, independently
of any religious considerations?
In the discussion on this question, the view was expressed by one of the invited speakers that, independent
of any belief in God or the immortality of the hum~n. s.o~l,
it was necessary to be concerned about the responslbllItles
of people to society. However, it was also su~gested that a
system of fully effective ethics should necessanly start from
COMPUTERS and AUTOMATION for November, 1960

or be based on a belief in God and the sacredness of the
human soul; otherwise it might limp.
2. Should a salesman for a computer manufacturer tell
all the truth and nothing but the truth?
3. If a computer manufacturer knows that his computer is going to become obsolete before very long, should
he say this to a prospect for a computer?
4. Do computer people have special responsibilities to
make known to non-computer people the powers of computers and their limitations? Are computer people especially responsible for avoiding exaggeration about computers
and data processors? In cases where different computer
people hold different views about the powers of computers,
should there be a way of settling these views, so that the
press and the public are not unduly disturbed?
It might be desirable to have a permanent public relations committee in the Association for Computing Machinery. The function of such a committee would be to help
members of the press and other non-computer people to
understand what computers can and cannot do, and over
the next five or ten years, mayor may not do - so that
alarmist reactions among readers of newspapers and other
magazines can be avoided.
5. What will be the social impacts of data processing
and computers on people who will be displaced from their
jobs as a result of computers? What can computer people
do to soften or modify these social impacts?
6. The Soviet Union is planning to use computers in
the operation of a planned economy, and this may lead to
considerable advantages for the Soviet Union. Should the


A 6" drum, 100 tracks, 150 heads, 3600 RPM,
stores 500,000 bits Manchester, overal1 size
10" x 12" with sealed cover. Designed to Military specifications. Similar to drum delivered to
Naval Ordnance Test Station. Some other customers include Link Aviation, Norden, Burroughs
Corp., Stanford Research Institute.

610 So. Arroyo Parkway

Pasadena, Calif.

MU rray 1-6721


United States make use of computers for auxiliary planning of the economy of the United States?
7. What is likely to be the impact of computers on industry as instruments of control here and there in the
management of industry?
8. Computers and data processors are being used in
intelligence operations. We can foresee a time when a vast
computing or data processing system will have information
on every person in the country, as a part of intelligence operations. This is likely to interfere with the personal privacy of individuals. What should we advocate in regard
to this question?
9. When a computer provides numbers, answers, or
decisions - "Multivac says ..." - does this lead to less
personal responsibility and personal judgment? Is there
a tendency to take refuge in the verdict of a computer in
regard to a difficult question?

I. From C. G. Stiefvater
Bridgeport, Conn., October 3, 1960

ions. If possible for you to do so, would you please tell us
if your organization has done any research in the following

1. The changes in commercial computers which are
anticipated in the coming years.
2. The future impact of different-size computers and
the market composition of such equipment.
3. The factors affecting the short-term and long-term
growth of computers.
We would appreciate any information which you could
provide in these areas. We will have to wrap up the study
by the week of October 10, and any prompt consideration
you can give to this matter would be quite beneficial.
II. From the Editor
Nearly everything which we find out is published in
the pages of Computers and Automation. We would suggest ,that you look through the last dozen issues of
Computers and Automation, and the indexes of the last
few years (usually published in the January issue), and
after you have done this, if you ask us any specific questions
which you may have, we shall be glad to try to answer them.

We are trying to ascertain what will be the future trends
or innovations in electronic data processing equipment. In
order to establish valid indicators, we are contacting knowledgeable people in this field to obtain their ideas or opin-

You can solve your computer prob.
lems quickly and economically by
using our 32K·word storage IBM
704. Whether you need long or short
runs, they can be readily scheduled
on our machine at the same attrac·
tive rate for every shift-$275 per
hour, including all peripheral
equipment and operators.
Bring your program and work in
our Client's Room between runsOr mail us your program with in·
structions for running it and we will
mail back the printout within 24
hours -- Or simply leave your
entire problem in the hands of our
capable mathematical analysts and
If you need pipe stress, structural
stress, flow analysis, or curve fitting,
one of our general programs might
well be just what you are look·
ing for.
To use our prompt, efficient com·
puter services, write or call us col·
lect, Hilltop 5·4321, extension 1449.



Computers and Automation is published 13 times a year
(monthly except 2 issues in June) at Boston, Mass.
1. The names and addresses of the publisher, editor, managing editor, and business manager are:
Publisher, Berkeley Enterprises, Inc., 815 Washington St.,
Newtonville 60, Mass.
Editor, managing editor, and business manager, Edmund
C. Berkeley, 34 Otis St., Newtonville 60, Mass.


The owner is: Berkeley Enterprises, Inc., 815 Washington St., Newtonville 60, Mass.
Stockholders holding one percent or more of the stock
Edmund C. Berkeley, 34 Otis St., Newtonville 60, Mass.
Max S. Weinstein, 25 Highland Drive, Albany 3, N.Y.

3. The known bondholders, mortgagees, and other security
holders o,wning or holding one percent or more of the total
amount of bonds, mortgages, or other securities are: None.
Edmund C. Berkeley, Editor
SWORN TO and subscribed before me, a notary public
in the Commonwealth of Massachusetts, on September 23,
Esther W. McHugh, Notary Public
My commission expires October 31, 1964.
COMPUTERS and AUTOMATION for November, 1960

Oct. 31-Nov. 2, 1960: 13th Annual Conference on Elec.
Techniques in Medicine and Biology, Sheraton Park
Hotel, Washington, D.C; contact G. N. Webb, Rm.
547, eSB, Johns Hopkins Hosp., Baltimore 5, Md.
Oct. 31 - Nov. 4, 1960: 7th Institute of Electronics in
Management, featuring Current Developments in Automatic Data Processing Systems, American University,
Washington, D.C; contact Dr. Lowell H. Hattery, Director, 7th Inst. on Electronics in Management, The
American University, 1901 F St., N. W., Washington
6, D.C
Nov. 5, 1960: Meeting of the Society for Industrial and
Applied Mathematics, Univ. of Pennsylvania, Physical
Sciences Bldg., 33rd & Chestnut Sts., Philadelphia, Pa.;
contact Dean Gillette, Bell Telephone Laboratories,
Inc., Whippany, N. J.
Nov. 15-17, 1960: Northeast Research & Engineering
Meeting (NEREM), Commonwealth Armory & Sheraton
Plaza Hotel, Boston, Mass.; contact J. H. Mulligan, Jr.,
Dept. of EE, NYU, New York 53, N.Y.

May 2-4, 1961: Electronic Components Conference, Jack
Tar Hotel, San Francisco, Calif.
May 7-8, 1961: 5th Midwest Symposium on Circuit
Theory, Univ. of 111., Urbana, Ill.; contact Prof. M. E.
Van Valkenburg, Dept. EE, Univ. of Illinois, Urbana,
May 9-11, 1961: Western Joint Computer Conference,
Ambassador Hotel, Los Angeles, Calif.; contact Dr. W.
F. Bauer, Ramo-Wooldridge Co., 8433 Fallbrook Ave.,
Canoga Park, Calif.
May 22-24, 1961:
Chicago, Ill.

National Telemetering Conference,

May 22-24, 1961: Fifth National Symposium on Global
Communications (GLOBECOM V), Hotel Sherman,
Chicago, Ill.; contact Donald C. Campbell, Tech. Program Comm., I.T.T. - Kellogg, 5959 S. Harlem Ave.,
Chicago 38, 111.

December 13-15, 1960: Eastern Joint Computer Conference, New Yorker Hotel, New York City; contact Dr.
Nathaniel Rochester, IBM, Yorktown Heights, N.Y.

June, 1961: Joint Automatic Control Conference, Univ. of
Colorado, Boulder, Colo.; contact Dr. Robert Kramer,
Elec. Sys. Lab., M.LT., Cambridge 39, Mass.

Jan. 16-19, 1961: ISA Winter Instrument-Automation Conference & Exhibit, conference at Sheraton-Jefferson Hotel,
exhibit at Kiel Auditorium, St. Louis, Mo.; contact William H. Kushnick, Exec. Dir., ISA, 313 Sixth Ave., Pittsburgh 22, Pa.

July 16-22, 1961: 4th International Conf. on Medical Electronics & 14th Conf. on Elec. Tech. in Med. & Bio., Waldorf Astoria Hotel, New York, N.Y.; contact Dr. Herman
P. Schwan, Univ. of Pa., School of Electrical Eng., Philadelphia, Pa.

Feb. 1-3, 1961: Winter Convention on Military Electronics,
featuring Communications, Telemetry, Data Handling
and Display, Los Angeles, Calif.; contact Dr. John J.
Meyers, Hoffman Electronics Corp., Military Products
Div., 3717 S. Grand Ave., Los Angeles 7, Calif.

Aug. 22-25, 1961: WESCON, San Francisco, Calif.; contact
Business Mgr., WESCON, 1435 La Cienega Blvd., Los
Angeles, Calif.

Feb. 15-17, 1961: International Solid State Circuits Conference, Univ. of Pa. and Sheraton Hotel Philadelphia,
Pa.; contact Jerome J. Suran, Bldg. 3, Rm. 115, General Electric Co., Syracuse, N. Y.
Mar. 20-23, 1961: IRE International Convention, Coliseum and Waldorf-Astoria Hotel, New York, N. Y.;
contact Dr. G. K. Neal, IRE, 1 E. 79 St., New York 21,
April, 1961: Joint Automatic Techniques Conference, Cincinnati, Ohio; contact J. E. Eiselein, RCA Victor Div.,
Bldg. 10-7, Camden 2, N.J.
Apr. 19-21, 1961: S. W. IRE Reg. Conf. and Elec.
Show, Dallas, Tex.; contact R. W. Olson, Texas Instruments Co., 6000 Lemmon Ave., Dallas 9, Tex.
COMPUTERS and AUTOMATION for November, 1960

Sept. 6-8, 1961: National Symposium on Space Elec. & Telemetry, Albuquerque, N.M.; contact Dr. B. L. Basore,
2405 Parsifal, N.E., Albuquerque, N.M.
Sept. 6-8, 1961: International Symposium on the Transmission and Processing of Information, Mass. Inst. of Technology, Cambridge, Mass.; contact Peter Elias, RLE, MIT,
Cambridge 39, Mass.
Sept. 6-8, 1961: 1961 Annual Meeting of the Association
for Computing Machinery, Statler Hotel, Los Angeles,
Calif.; contact Benjamin Handy, Chairman Local Arrangements Committee, Litton Industries, Inc., 11728 W.
Olympic Blvd., W. Los Angeles, Calif.
Sept. 11-15, 1961: The Third International Congress on
Cybernetics, Namur, Belgium; contact Secretariat of The
International Association for Cybernetics, 13, rue Basse
Marcelle, Namur, Belgium.

Make over 200 Small
Computing and Reasoning
Machines with ...


WHAT COMES WITH YOUR BRAINIAC® KIT? All 33 experiments from our original Geniac® kit (1955), with
exact wiring templates for each one. All 13 experiments from the former Tyniac kit. 156 entirely new experiments
(Brainiacs) with their solutions. Over 600 parrs, as follows: 6 Multiple Switch Discs; Mounting Panel; 10 Flashlight
Bulbs; 2 Multiple Socket Parts, each holding 5 bulbs; 116 Wipers, for making good electrical contact (novel design,
patented, no. 2848568); 70 Jumpers, for transfer contacts; 50 feet of Insulated Wire; Flashlight Battery; Battery Box;
nuts, bolts, sponge rubber washers, hard washers, screwdriver, spintite blade, etc. ALSO: 256 page book, "Brainiacs"
by Edmund C. Berkeley, including chapters on: an introduction to Boolean Algebra for designing circuits; "How to
go from Brainiacs and Geniacs to Automatic Computers"; complete descriptions of 201 experiments and machines; over
160 circuit diagrams; list of references to computer literature.
This kit is an up-to-the-minute introduction to the design of arithmetical, logical, reasoning, computing, puzzle-solving,
and game-playing circuits - for boys, students, schools, colleges, designers. It is simple enough for intelligent boys to assemble, and yet it is instructive even to engineers, because it shows how many kinds of computing and reasoning circuits
can be made from simple components. This kit is the outcome of 11 years of design and development work with Geniacs
and small robots by Berkeley Enterprises, Inc. With this kit and manual you can easily make over 200 small electric brain
machines that display intelligent behavior and teach understanding first-hand. Each one runs on one flashlight battery; all
connections with nuts and bolts; no soldering required. (Returnable for full refund if not satisfactory.) .. Price $18.95
Syllogism Prover
James McCarty's Logic Machine
IF, NEITHER ... NOR Machines
A Simple Kalin-Burkhart Logical Truth Calculator
The Magazine Editor's Argument
The Rule About Semicolons and Commas
The Farnsworth Car Pool
Black Match
Sundorra 21
Frank McChesney's Wheeled Bandit
COMPUTERS - to add, subtract, multiply, divide, . . . • ,
using decimal or binary numbers.
- to convert from decimal to other scales of notation and
vice versa, etc.
Operating with Infinity
Adding Indefinite Quantities
Factoring Any Number from 45 to 60
Prime Number Indicator for Numbers 1 to 100
Thirty Days Hath September
Three Day Weekend for Christmas
Calendar Good for Forty Years 1950 to 1989
Money Changing Machine
Four by Four Magic Square
Character of Roots of a Quadratic
Ten Basic Formulas of Integration
The Missionaries and the Cannibals
The Daisy Petal Machine
Calvin's Eenie Meenie Minie Moe Machine
The Cider Pouring Problem
The Mysterious Multiples of 76923, of 369, etc.
Bruce Campbell's Will
The Fox, Hen, Corn, and Hired Man
The Uranium Shipment and the Space Pirates
General Alarm at the Fortress of Dreadeerie
The Two Suspicious Husbands at Great North Bay

The Submarine Rescue Chamber Squalux
The Three Monkeys Who Spurned Evil
Signals on the Mango Blossom Special
The Automatic Elevator in Hoboken
Timothy's Mink Traps
Josephine's Man Trap
Douglas Macdonald's Will
Word Puzzle with TRICK
The Waxing and the Waning Moon
Intelligence Test
Guessing Helen's Age
Geography Quiz
Mr. Hardstone's Grammar Test
Solving Right Triangles
The Jiminy Soap Advertising SIgn
The Sign that Spells Alice
Tom, Dick, and Harry's Private Signaling Channels
Jim's and Ed's Intercom
Secret Coder
Secret Decoder
Lock with 65,000 Combinations
Lock with 15,000,000 Combinations
The General Combination Lock
Leonard's Two-Way Coding Machine


or a copy 0/ it ..

Berkeley Enterprises, Inc.
815 Washington Street, R149, Newtonville 60, Mass.
Please send me BRAINIAC KIT K18, including manual,
instructions, over 600 parts, templates, circuit diagrams, etc.
I enclose $18.95 for the kit plus ............. for handling and
shipping (30c, east of Mississippi; 80c, west of Mississippi;
$1.80, outside U.S.). I understand the kit is returnable in
seven days for full refund if not satisfactory (if in good
condition) .
My name and address are attached.
COMPUTERS and AUTOMATION for November, 1960

Moses M. Berlin
Cambridge, Mass.
E PUBLISH HERE citations
and brief reviews of books
and other publications which have a
significant relation to computers, data
processing, and automation, and
which have come to our attention.
\Y/ e shall be glad to report other information in future lists if a review
copy is sent to us. The plan of each
entry is: author or editor / title /
publisher or issuer / date, publication
process, number of pages, price or
its equivalent / comments. If you
write to a publisher or issuer, we
would appreciate your mentioning
Computers and Automation.


Ulam S. M. / A Collection of Mathematical Problems / Interscience Publishers. Inc., 250 Fifth Ave., New York
1, N. Y. / 1960, printed, 150 pp, $5.00
This is an advanced book stating problems from the following branches of
mathematics: Set Theory, including infinite sets and projective algebras; Algebra;
Metric Spaces; Topological Spaces; Topological Groups; and Analysis; Physical
Systems; and, Computing Machines as a
HeuristIc Aid. "The problems listed arc
regarded as unsolved in the sense that
the author does not know the answers."
The author reports on using computing
machines for checking intelligent guesses
in mathematical physics, combinatorial
analysis, number theory, etc. A bibliography is included.

Young, J. Z. / Doubt and Certainty in
Science - A Biologist's Reflections on
the Brain / Galaxy Book Series, Oxford University Press, New York,
N. Y. / 1960, printed, 168 pp, $1.50

Flagle, Charles D., William H. Huggins,
Robert H. Roy / Operations Research
and Systems Engineering / The Johns
Hopkins Press, Baltimore 18, Md. /
1960, printed, 889 pp, $14.50.

A part of this book, first published in
1951, is devoted to a discussion of memory stores - of the human being and of
the machine. The author couples this part
with some opinions on the framework of
society, and the interconnection between
the human nervous system and the electronic system of computers. The text was
first presented as eight lectures; the book
also contains a list of references and an

This book is a compendium on the
meaning, use, and importance of the fields
of operations research and systems engineering. The first part of the book, "Perspectives," includes seven papers which discuss
the background and history relating to the
development of these fields; part two,
"Methodologies," contains sixteen papers including, "Simplified Models in Operations
Research," "Electronic Digital Computers,"
and, "System Dynamics"; part three, "Case
Studies," contains four papers and includes
"Simulation of Tactical War Games."

Diamond, Solomon / Information and Error / Basic Books, Inc., 59 Fourth Ave.,
New York 3, N. Y. / 1960, printed,
307 pp, $5.00
The aim of this book is to provide a
student of psychology who does not have
much mathematical preparation with information about the level of statistics
which is of importance in the study of
psychology. The text, with examples, analogies, style, and humor, serves as an
introduction to statistical analysis, including, in thirteen chapters, variance, probability, proportions, chi-square, product
moment correlation, factor analysis and
a discussion on handling nonnormal data,
etc. An appendix includes tables. A glossary of $ymbols, a list of formulas, and an
index are included.
Alt, Franz L., editor; Calvin Gotlieb,
N. A. Phillips, Y. Bar-Hillel, A. C.
Samuel, R. Fatehchand, G. W. Reitwiesner, contributors / Advances in
Computers / Academic Press, Inc., 111
Fifth Ave., New York 3, N. Y. / 1960,
printed, 316 pp, $10.00.
This is the first volume of a series reporting progress; this book contains six important, full surveys on the following topics: General-Purpose Programming for Business Applications; Numerical Weather Prediction; The Present Status of Automatic
Translation of Languages; Programming
Computers to Play Games; Machine Recognition of Spoken Words; and Binary Arithmetic. Lists of references, subject index,
and name index are included_

Reifler, Erwin, W. Ryland Hill, David L.
Johnson, and others / Linguistic and
Engineering Studies in Automatic Language Translation of Scientific Russian
into English / University of Washington Press, Seattle 5, Wash. / 1960,
printed, 658 pp., $10.00

Information Processing: Proceedings of
the International Conference on Information Processing Unesco, Paris, 15-20
June 1959 / UNESCO Publications
Center, 801 Third Avenue., New York
22, N. Y. / 1960, printed, 520 pp,

A summary of the fundamental problems, procedures and achievements of a
lexicographical research project to establish an automatic system for machine
transla60n other than the electronic computer systems previously used, is presented. In two parts, Linguistic Analysis and
Engineering Analysis, the report includes:
an outline of the project; the selected
Russian texts used; the use of the IBM
650 computer for the study of syntax;
dictionary card processing procedures;
and five appendices consisting mainly of
research papers. Two of the appendices
are entitled "A Coding and Operational
Program for Machine Translation Using
a High-Capacity Optical Memory", and
"The Design of a Practical Russian-English Mechanical Translator".

This large book (12" by 9", in small
print) publishes the complete proceedings
of the conference, giving 61 papers (in
English or French), preceded by summaries
in English, French, German, Russian, and
Spanish, and completed by reports of the
discussion. In addition, there are summaries
in English or French, of the 65 lectures
given at various specialized meetings. The
seven divisions are: Methods of Digital
Computers, Common Symbolic Language
for Computers, Automatic Translation of
Languages, Pattern Recognition and Machine Learning, Logical Design of Computers, Special Session on Computer Techniques of the Future, and Miscellaneous
Topics. A subject index and a list of participants are included.

COMPUTERS and AUTOMATION for November, 1960

!vall, T. E. / Electronic Computers Principles and Applications, Second
Edition / Philosophical Library, Inc.,
15 East 40 St., New York 16, N. Y. /
1960, printed, 263 pp, $15.00.
The major part of this book is devoted
to discussions and descriptions of the circuitry and construction of digital and analog
computers. Following the discussions, applications are enumerated. Among the fourteen chapters are: "Evolution of the Computer," "General Principles of Computing,"
"Recent Developments," and "Computers of
the Future," Index.
Booth, Andrew D., editor / Progress in
Automation / Academic Press, Inc.,
111 Fifth Ave., New York 3, N. Y. /
1960, printed, 231 pp, $8.50.
In this first volume of a series on automation in Great Britain, an introduction by
the editor and ten papers on related topics
are given. Among the papers, are: "Analogue-to-Digital Conversion Techniques,"
"Application of Electronics to Process Control Systems," "The Ferranti System of Machine Tool Control," and, "The E. M. I.
System of Machine Tool Control." A name
index and subject index are included.
Langer, Rudolph E., editor / Frontiers
of Numerical Mathematics / The University of Wisconsin Press, 811 State
St., Madison, Wisconsin / 1960, photooffset, 132 pp, $3.50.
Eight invited papers were delivered at a
Symposium conducted by the Mathematics
Res. Center, U. S. Army, and the Nat'l.
Bureau of Standards. The objective of this
symposium was to survey the future and
identify some of the mathematical problems
which will have to be faced in the lines of
scientific advance. The eight speakers were
asked to outline the impending mathematical tasks as they saw them. Eighteen
invited discussants also took part in the
symposium, and the discussions of the papers are also reported.
The titles of the papers and the authors
are: Stress Analysis in the Plastic Range,
William Prager; Some Mathematical Problems of Nuclear Reactor Theory, Garrett
Birkhoff; Numerical Problems of Contemporary Celestial Mechanics, Zdenek Kopal;
Aeroelasticity, Lee Arnold; Operations Research, Philip M. Morse; Mathematical Bottlenecks in Theoretical Chemistry, J. O.
Hirschfelder; Magnetohydrodynamics, S.



Chandrasekhar, The Solution of Systems of
Partial Differential Equations Arising in
Meteorology, J. Smagorinsky.


Vajda, S. / An Introduction to Linear
Programming and the Theory of
Games / John Wiley & Sons, Inc., 440
Fourth Ave., New York 16, N. Y. /
1960, printed, 76 pp, $2.25.

6th Annual Issue,
the only directory in the
computer field,
a regular issue of
Computers and Automation

Part 1, ROSTER OF ORGANIZATIONS. Each entry gives:
Name of organization / Address
/ Telephone number / Types of
computers, data processors, accessories, components, services, etc., produced or offered / Approximate
number of employees / Year organization was established
(over 700 organization entries)
SERVICES. Each expanded boldface entry gives:
Name or identification of product
or service / Brief description (20 to
50 words) / Uses / Price range, between ... and ...
Other entries are cross-references.
(Over 2000 product and service
entries in total)
This is your indispensable guide
to organizations, products, and services in the computer field - useful
to you the whole year.
Send for your own copy now
(or a copy of it)
To: Computers and Automation
815 Washington St., R149
N ewtonviUe 60, Mass.
Please send us a copy of the 1960
Computer Directory. We enclose
$6.00. Returnable in 7 days for full
refund if not satisfactory.
Name ................................................. .
Title ................................................... .
Organization ..................................... .
Address ............................................... .


Part I of this book includes seven chapters
on linear programming - an introduction,
information about graphical representation,
the simplex method, duality, etc. Appendix
I presents a proof of the Fundamental
Theorem of Duality. Part II of the book
on Game Theory includes short chapters on
fundamental concepts, graphical representation, non-zero-sum games, and infinite
games. Appendix II presents a proof of the
so-called Main Theorem of the Theory of
Games. Index.
Haley, A. C. D., and W. E. Scott, editors,
and 7 more authors / Analogue and
Digital Computers / Philosophical Library, Inc., 15 East 40 St., New York
16, N. Y. / 1960, printed, 308 pp,
This book presents basic information on
the design and construction of analog and
digital computing systems. The ten chapters
include: Introduction to Computers; Operation and Applications of Analogue Computers; Number Representation in Digital
Computers; Operation of a Digital Computer; Circuit Elements and Computer
Units; Storage; Programming. Index.
Bibbero, Robert J. / Dictionary of Automatic Control/Reinhold Pub. Corp.,
430 Park Ave., New York 22, N. Y. /
1960, printed, 282 pp, $6.00.
This dictionary of automatic control terminology includes, besides definitions, a condensed discussion of each term. The subjects included are: control theory and basic
concepts; computers and data processing;
industrial machine and process control; aircraft and missile control and telemetering;
and control components and design factors.
A classified index supplements the regular
alphabetic listing, classifying terms under
the five main subjects.
Felix, Lucienne / The Modern Aspect of
Mathematics / Basic Books, Inc., 59
Fourth Ave., New York 3, N. Y. /
1960, printed, 194 pp, $5.00.
The "revolution" in mathematics, led by
the fictional Nicolas Bourbaki, is chronicled
in this book. The authoress, who is in close
contact with the group of Frenchmen involved in the reorganization of mathematics - from general algebra to topology
- has compiled six chapters on the new
trends in, and the new pedagogic point of
view on, the subject. A glossary of symbols,
two appendices and an index are included.
Journal of Mathematical Analysis and
Applications / Academic Press, Inc.,
111 Fifth Ave., New York 3, N. Y. /
1960, (vol. I), printed, $16.00.
This publication, designed to provide a
medium for the rapid publication of mathematical papers, will include papers devoted
to the mathematical treatment of questions

ansmg in physics, chemistry, biology, and
engineering. To minimize delays between
receipt and publication, each of the Associate Editors may accept manuscripts. Among
the Associates are, N. Levinson, M. 1. T.,
G. Birkhoff, Harvard Univ., H. N. Shapiro,
N. Y. U., and L. Zadeh, U. of Cal., Berkeley.
Dr. R. Bellman, RAND Corp., serves as
the Editor.
Forsythe, George E., and Wolfgang R.
Wasow / Finite-Difference Methods
for Partial Differential Equations /
John Wiley & Sons, Inc., 440 Fourth
Avenue., New York 16, N. Y. / 1960,
printed, 444 pp, $11.50.
This book concentrates on the more important finite-difference methods for solving
partial differential equations, including initial-value and boundary-value problems, and
other topics relevant to the solution of
problems using computers. The contents
include: Introduction to Partial Differential
Equations and Computers; Hyperbolic Equations in Two Independent Variables; Parabolic Equations; Elliptic Equations; InitialValue Problems in more than Two Independent Variables; Index; and a comprehensive Bibliography. The major topics are
subdivided into many sections. A graduate
course in partial differential equations is
not necessary to the understanding of this
book, but "a good course in advanced calculus and some knowledge of matrix theory"
is advocated for understanding the book.
This book is intended primarily for persons
wishing to understand the numerical analysis underlying the use of difference methods.
Kells, Lyman M. / Elementary Differential Equations, 5th Edition, / McGrawHill Book Co., Inc., 330 West 42 St.,
New York 36, N. Y. / 1960, printed,
318 pp, $6.25.
From an introduction to the subject, including definitions and elementary problems, to a final chapter on applications of
partial differential equations, this book discusses in detail and with a care for rigor,
the major topics, including: first order equations, Laplace transforms, solutions by series, and partial differential equations.
Answers to the problems in each chapter,
are given after the text. An index is included.
Jonscher, A. K., / Principles of Semiconductor Device Operation / John
Wiley & Sons, Inc., 440 Fourth Ave.,
New York 16, N. Y. / 1960, printed,
168 pp, $5.00.
A discussion is presented, of injection,
decay, and transport of "excess carrier,"
densities in semiconductors. The six chapters are: Outline of Semiconductor Theory,
Non-Equilibrium Carrier Densities, Transport of Excess Carrier Densities in a Homogeneous Medium, Theory of p-n Junctions and Junction Diodes, Theory of MultiJunction Structures, and Carrier Flow in
Inhomogeneous Media. Eight appendices
contain information about various topics including vector operators, and the error function complement; a list of symbols follows
the text. Index. The author is at General
Electric Co., Ltd., Research Laboratories,
Wembley, England.

COMPUTERS. and AUTOMATION for November, 1960

Essential Special Terms in Computers and Data
Processing - Suggested List, and Definitions
Part 3
(From: Glossary of Terms in Computers and Data Processing, 5th edition
of the Computers and Automation
glossary / Edmund C. Berkeley and
Linda L. Lovett / Berkeley Enterprises,
Inc., 815 Washington St., Newtonville
6:J, Mass. / July 1960,96 pages, photooffset, $3.95 to nonsubscribers, $1.85
to subscribers.)
VI. Operation
check digit(s) - One or more digits carried along with a machine word (i.e.,
a unit item of information handled by
the machine) , which report information
about the other digits in the word in
such fashion that if a single error occurs
(excluding two compensating errors),
the check will fail and give rise to an
error alarm signal. For example, the
check digit may be 0 if the sum of other
digits in the word is odd, and the check
digit may be 1 if the sum of other digits
in the word is even. It is possible to
choose check digits for rows and columns
in a block of characters recorded on magnetic tape, for example, in such a way
that any single error of a 1 for a 0 or
a 0 for aI, can be located automatically
by row and column, and eliminated
automatically by the computer.

hours of scheduled computer operation
including time when the machine is undergoing preventive maintenance. Same
as "operating ratio," which see.
on-line (adjective) - Equipment. Operating in time with and under the direct
control of the central equipment. If the
central equipment is an automatic computer, and if the automatic computer is
not operating in real time, then on-line
operation does not imply real-time operation. But if a computer is part of another process taking place in real-time,
then on-line operation implies real-time
operation. For example, data from wind
tunnel experiments may be fed into the
computer directly from observing instruments, and the computer may report results of the experiment within a few
seconds from the time the experiment is
finished: this is on-line operation.
VII. Representation of Information
digit - 1. One of the symbols 0, 1, 2, 3,
4, 5, 6, 7, 8, 9, used in numbering in
the scale of ten. 2. One of these symbols and sometimes also letters expressing integral values ranging from 0 to
n-1 inclusive, used in a scale of numbering to the base n.

of a chosen base number n (sometimes
more than one). If a quantity is written
in the scale of notation n, then the successive positions of the digits report the
powers of n. Thus 379 in the scale of
10 or decimal notation means 3 hundreds, 7 tens, and 9. The number 379
in the scale of 16 (used in some computers) means 3 times sixteen squared,
plus 7 times sixteen, plus 9 (which in
decimal notation would be 889). 1101
in the scale of two means 1 eight, 1 four,
o twos and lone (which in decimal
notation would be 13). In writing numbers, the base may be indicated by a
subscript (expressed always in decimal
notation) when there may be doubt
about what base is employed. For example, 11.1012 means two, plus one,
plus one half, plus one eighth, but
11.1013 means three plus one, plus one
third, plus one twenty-seventh. Names
of scales of notation which have had
some significant consideration are:

2, 5

quaternary, tetra I
hexadecimal, sexidecimal

automatic checking - Computers. Provision, constructed in hardware, for automatically verifying the information,
transmitted, manipulated or stored by
any device or unit of the computer.
Automatic checking is "complete" when
every process in the machine is automatically checked; otherwise it is partial.
The term "extent of automatic checking"
means either (1) the relative proportion
of machine processes which are checked,
or (2) the relative proportion of machine hardware devoted to checking.

character - Digital Computers. 1. A decimal digit 0 to 9, or a letter A to Z,
either capital or lower case, or a punctuation symbol, or any other single symbol (such as appear on the keys of a
typewriter) which a machine may take
in, store, or put out. 2. One of a set
of basic or elementary unit symbols
which, singly or in sequences of two
or more, may express information and
which a computer may accept. 3. A
representation of such a symbol in a
pattern of ones and zeros representing
a pattern of positive and negative pulses
or states.

computing efficiency - Computer Operation. The ratio obtained by dividing (1)
the total number of hours of correct machine operation (including time when
the program is incorrect through human
mistakes) by ( 2) the total number of

notation (in the sense "scale of notation"
or "positional notation" for numbers)
- Arithmetic. A systematic method for
stating quantities in which any number
is represented by a sum of coefficients
times multiples of the successive powers

binary notation - The writing of numbers
in the scale of two. Positional notation
for numbers using the base 2. The first
dozen numbers zero to eleven are written in binary notation as 0, 1, 10, 11,
100, 101, 110, 111, 1000, 1001, 1010,
1011. The positions of the digits designate powers of two; thus 1010 means 1
times two cubed or eight, 0 times two
squared or four, 1 times two to the first
power or two, and 0 times two to the
zero power or one; this is equal to one
eight plus no four's plus one two plus
no ones, which is ten.

ARTICLES : We desire to publish articles
that are factual, useful, understandable, and
interesting to many kinds of people engaged
in one part or another of the field of computers and automation. In this audience are
many people who have expert knowledge of
some part of the field, but who are laymen
in other parts of it.
Consequently, a writer should seek to explain his subject, and show its context and
significance. He should define unfamiliar
terms, or use them in a way that makes their
meaning unmistakable. Ordinarily, the
length should be 1000 to 3000 words. A

suggestion for an article should be submitted to us before too much work is done.
PAYMENTS: In many cases, we make small
token payments for articles, if the author
wishes to be paid. The rate is ordinarily
Yzc a word, the maximum is $15, and both
depend on length in words, whether printed
before, etc.
All suggestions, manuscripts, and inquirdressed to: The Editor, COMPUTERS and
ies about editorial material should be adAUTOMATION, 815 Washington Street,
Newtonville 60, Mass.

WE ARE interested in articles,
papers, reference information, and
discussion relating to computers and
automation. To be considered for
any particular issue, the manuscript
should be in our hands by the first
of the preceding month.

COMPUTERS and AUTOMATION for November, 1960

The digits used for "ten" and "eleven"
are ordinarily "t" and "e"; beyond eleven,
uniformity of nomenclature has apparently not yet developed.




Basic Source Information,
Available to You from

24 "issues" per year,
distributed 13 times a year
Mailing plates for over 22,000 computer people. We address your envelopes.
$19.00 per M
The Computer Directory and Buyers'
Guide, 1960 (the June, 1960, issue of
over 700 organization listings, over 2000
product and service listings.
Over 860 careful, clear, understandable
definitions. 5th cumulative edition. (20
or more copies, 10% discount.)
(For nine years of publication.)
If available, $1.25 each, except Directory issues June 1955 to 1959, $4.00
each; June, 1960, $6.00.
U.S.A. one year $7.50; two years $14.50;
add SOc per year for Canada, $1.00 per
year elsewhere.
These rates apply to prepaid subscriptions to COMPUTERS and AUTOMATION coming in together direct to the

( United States)
Number of

7 or more


Rate for Each
Subscription, and
Resulting Saving:
One Year
Two Years

6.30 - 16
6.80 - 9
7.10- 5

11.00 - 24
12.20 - 16
13.20- 9

For Canada, add 50 cents for each year;
outside of the United States and Canada,
add $1.00 for each year.

Send prepaid orders or requests for
more information to:
815 Washington St., Newtonville 60, Mass.
If not satisfactory, returnable in seven days
for full refund.


coded decimal (adjective) - Computers. A
form of notation by which each decimal
digit separately is converted into a pattern of binary ones and zeros. For example, in the "8-4-2-1" coded decimal
notation, the number twelve is represented as 0001 0010 (for 1, 2) whereas
in pure binary notation it is represented
as 1100. Other coded decimal notations
are known as: "5-4-2-1," "excess three,"
"2-4-2-1," etc. Following are the codes
for the decimal digits 0 to 9 in each of
the mentioned systems:
Decimal 8-4-2-1 5-4-2-1 three 2-4-2-1








biquinary notation - Numbers. A scale of
notation in which the base is alternately
2 and 5. For example, the number 3671
in decimal notation is 03 11 12 01 in
biquinary notation; the first of each pair
of digits counts 0 or 1 units of five, and
the second counts 0, 1, 2, 3, or 4 units.
Roman numerals are essentially a biquinary notation, except that different
letters are used in each place, V and I
in the first place, X and L in the second
place, C and D in the third, etc.; for
example, the biquinary number 03 11
12 01 is in Roman numerals MMMDCLXXI. Biquinary notation expresses
the representation of numbers by the
abacus. and by the two hands and five
fingers of man: and has been used in
some automatic computers.
binary digit - A digit in the binary scale
of notation. This digit may be only 0
(zero) or 1 ( one). It is equivalent to
an "on" condition or an "off" condition,
a "yes" or a "no," etc.
bit - A binary digit; a smallest unit of
information; a "yes" or a "no"; a single
pulse in a group of pulses; a single magnetically polarized spot in a group of
such spots. This word is derived from
the "b" in "binary" and the "it" in
"digit"; the word replaces the obsolete
blend word "bigit," and takes on added
meaning from the word "bit" meaning
"small piece."
machine language - Computers. 1. Information in the physical form which a
computer can handle. For example,
punched paper tape is machine language, while printed characters on paper
are not usually machine language. 2.
Numbers or instructions expressed in a
form that a computer can process at
once without conversion, translation, or
programmed interpretation. Note that a
punched card containing an instruction
in punched holes that requires programmed interpretation is "in machine
language" by the first meaning and is

"not in machine language" by the second meaning.
machine word - Digital Computers. A
unit of information of a standard number of characters, which a machine regularly handles in each transfer. For example, a machine may regularly handle
numbers or instructions in units of 36
binary digits: this is then the "machine
word." See also "information word."
VIII. Mathematics and Logic
fixed-point calculation - Computers. Calculation using or assuming a fixed or
constant location of the decimal point
or the binary point in each number.
floating-point calculation Computers.
Calculation taking into account varying
location of the decimal point (if base
10) or binary point (if base 2), and
consisting of writing each number by
specifying separately its sign, its coefficient, and its exponent affecting the base.
For example, in floating-point calculation, the decimal number -638,020,000
might be reported as -,6.3802,8, since
it is equal to -6.3802 x 10 8 •
complement - 1. Arithmetic. A quantity
which is derived from a given quantity,
expressed in notation to the base n, by
one of the following rules. (a) Complement on n: subtract each digit of the
given quantity from n-1, add unity to
the rightmost digit, not zero and perform all resultant carries. For example,
the twos complement of binary 11010
is 00110; the twos complement of 0001
1010 is 11100110; the tens complement
of decimal 679 is 321; the tens complement of 000679 is 999321. (b) Complement on n-1: subtract each digit of
the given quantity from n-1. For example, the ones complement of binary
11010 is 00101; the ones complement of
00011010 is 11100101; the nines complement of decimal 679 is 320; the nines
complement of 000679 is 999320. The
complement is frequently employed in
computers to represent the negative of
the given quantity. 2. Boolean Algebra.
The element equal to the universe element except the stated element; the
result of the operation NOT... or ALL
EXCEPT. . . The complement of a
Boolean element a is NOT-a, or a', or
'-' a.
parameter - 1. Mathematics. A constant
or variable which enters fundamentally
into a mathematical function and which
has the property that its different values
produce different functions. For example, the function y = ax + b has two
parameters, a and b; when a and bare
constant, y = ax + b represents a line,
but the choice of values of a and b
determines the angles and distances at
which the line cuts the coordinate axes.
2. Digital Computer Programming. In
a subroutine, a quantity which may be
given different values when the subroutine is used in different parts of one
main routine, but which usually remains
unchanged throughout anyone such usc.
For example, a parameter may specify

COMPUTERS and AUTOMATION for November, 1960


the number of characters in an item,
the position of the decimal point, the
number of columns in a field, the number of times a certain cycle of operations
is to be repeated, etc. To use a subroutine successfully in many different
programs requires that the subroutine
be adaptable by changing its parameters.

From time to time we bring up to
date our "Who's Who in the Computer Field." We are currently asking all computer people to fill in the
following Who's Who Entry Form,
and send it to us for their free listing
in the Who's Who that we publish
from time to time in Computers and
Automation. Weare often asked
questions about computer people and if we have up to date information in our file, we can answer those
If you are interested in the computer field, please fill in and send us
the following Who's Who Entry
Form (to avoid tearing the magazine, the form may be copied on any
piece of paper).
Name? (please print)
Your Address?
Your Organization?
Its Address? ....... .
Your Title? ....... .
Your Main Computer Interests?
) Applications
) Business
) Construction
) Design
( ) Electronics
( ) Logic
( ) Mathematics
( ) Programming
( ) Sales
( ) Other (specify):

inclusive OR - Logic. A logical operator
which has the property that if P and Q
are two statements, then the statement P
or Q is true if and only if P is true or
if Q is true or if both P and Q are true.
See also under "OR."

Year of birth? ..... . ..................... .
College or last school? ...... ............ .
Year entered the computer field? ..... .
Occu pation? .................................... .
Anything else? (publications, distinct ions, etc.) ...... ...... ........................ .

NOT - Logic. A logical operator that has
the property that if P is a statement,
then the statement "NOT-P" ("it is not
the case that P") , is true if the statement
P is false, and false if the statement P is
true. The NOT operator is often represented as follows: P' (read "P prime"),
P (read "P dash"), or '"""' P (read
"tilde P").



Boolean algebra - An algebra like ordinary algebra but dealing instead with
classes, propositions, on-off circuit elements, etc., associated by operators AND,
etc., and permitting computations and
demonstration, as in any mathematical
system, making use of symbols efficient
in calculation. This algebra was named
after George Boole, famous English
mathematician (1815-1864).

Computers and Automation's Fifth
Edition (1960) of the:

96 pages long, this edition contains
over 860 computer terms and expressions with their definitions, EXPLAINED so that people new to the
computer field can understand them.
(Our lase edition, October, 1956, contained 490 terms.) This will be an
invaluable guide to "understanding
your way around" the computer field.
Returnable for full refund within 10
days if not satisfactory.

AND - 1. Logic (and Boolean Algebra).
A logical (or Boolean algebra) operator which has the property that if P
and Q are two statements, then the
statement "P AND Q" is true or false
precisely according to the following table
of possible combinations:




Each monthly issue of Computers
and Automation now has 12 additional pages of "News of Computers
and Data Processors: ACROSS THE
EDITOR'S DESK." Of course, articles,
ideas, forum, reference information,
surveys, etc., are all there still; and the
June issue is the annual edition of
subscription rate, U.S.A., one year,
FER: both our new Glossary and
Computers and Automation for one
year, $7.90. You save 30%.

The AND operator is often represented
by a centered dot (-), or by no sign,
as in P-Q, PQ. 2. Circuits. A connection between two circuits A and B or
two circuit elements A and B which
passes a signal if and only if both A
and B contain the signal.


exclusive OR - Logic. A logical operator
that has the property that if P and Q
are two statements, then the statement
P OR ELSE Q is true precisely according
to the following table of possible combinations:

When you have filled in this entry
form please send it to: Who's Who
Editor, Computers and Automation,
815 Washington Street, Newtonville
60, Mass.





The exclusive OR operator, the OR
ELSE operator, has the property: P OR
ELSE Q is equivalent to P AND NOT
Q, OR Q AND NOT P, and accordingly
may be written in symbols P-Q' v P'-Q.

COMPUTERS and AUTOMATION for November, 1960

- Mail this coupon, or a copy of it To: COMPUTERS AND AUTOMA.
TION, 815 Washington St., R 149,
Newtonville 60, Mass.
o Please send me the new Glossary.
I have checked the arrangement
which applies:
Subscriber Already
Returnable for full refund within 10 days if not satisfactory.
Please send me the special combination offer of Computers and
Automation for one year and
the Glossary when printed. I
enclose $7.90.
o Other requests (please explain)



My name and address are attached.


Reg. Patent Agent
Ford lnst. Co., Div. of Sperry Rand Corp.
Long Island City 1, New York

follow ing is a compilation of
patents pertaining to computers
and associated equipment from the
"Official Gazette of the United
States Patent Office," dates of issue
as indicated. Each entry consists of:
patent number I inventor (s ) I assignee I invention. Printed copies
of patents may be obtained from the
U.S. Commissioner of Patents, Washington 25, D.C, at a cost of 25
cents each.
May 10, 1960 (Cont'd)
2,936,116 I Phil. A. Adamson, San Gabriel,
and Howard L. Engel and Eldred C. Nelson, Los Angeles, Calif. I Hughes Aircraft Co., a corp. of Del. I An electronic
digital computer.
2,936,117 / Elmer L. Younker, Madison,
N.]. / Bell Telephone Lab., Inc., New
York, N.Y. / A high speed switching
circuit employing slow acting components.
2,936,118 I George B. Greene, Berkeley,
Charles M. Hill, Piedmont, Eugene P.
Hamilton, Richmond, and William B.
Bennett, Berkeley, Calif., and William
P. Steed, Philadelphia, Pa. I Marchant
Research, Inc., a corp. of Calif., I An
electronic digital computer.
2,936,119 / Dana M. Collier and Leighton
A. Meeks, Oak Ridge, Tenn., and James
P. Palmer, Stony Brook, N.Y. / U.S.A.
as represented by the U.S. Atomic Energy
Comm. I A simultaneous differential
equation computer.
2,936,120 I Samuel D. Bedrosian, Havertown, and Frank P. Simmons, Devon,
Pa. / Burroughs Corp., Detroit, Mich. /
A great circle and celestial data computer.
May 17, 1960
2,936,951 I Austin J. Maher, Jr., Brooklyn.
N.Y. / Sperry Rand Corp., Ford Inst.

Ca., Div., a corp of Del. / A method
and apparatus for accurate analog integration of time functions.
2,936,956 / Martin Kassel, Berlin - Charlottenburg, Germany / Kienzle Apparate
G.m.b.H., Villingen / Schwarzwald,
Germany / An electronic computer for
converting single electrical pulses into
pulse sequences each composed of a
predeterminable number of pulses.
2,937,288 / Frederic C. Williams and
George B. Chaplin, Romiley, Eng. /
National Research Dev. Corp., London,
Eng. / A shift register circuit.
2,937,289 / Richard W. Aldrich, Liverpool,
and Jerome J. Suran, Syracuse, N.Y. /
General Elec. Co., a corp. of N.Y. / A
d,gital to analog converter.
2,937,290 / Kan Chen, Wilkinsburg, Pa. /
Westinghouse Electric Corp., East Pittsburgh, Pa. / An anti-coincident circuit.
2,937,291 / Leonard R. Harper, Poughkeepsie, N.Y. / I.B.M. Corp., New York,
N.Y. / A single shot bistable circuit.
2,937,363 / Robert 1. Roth, Mount Pleasant, N.Y. / 1.B.M. Corp., New York,
N.Y. / A data processing machine.
2,937,364 / Milton Rosenberg, Santa Monica, Calif. / Telemeter Magnetics, Inc.,
a corp. of Calif. / A magnetic memory
2,937,366 / John C. Sims, Jr., Springhouse,
Pa., / Sperry Rand Corp., a corp. of
Del. / A pulse group synchronizer.
May 24, 1960
2,937,810 / William A. Wadsworth, Towaco, N.J. / Bell Telephone Lab., Inc.,
New York, N.Y. / A serial digital computer circuit for summing successive increments.
2,938,193 / John P. Eckert, Jr., Gladwyne,
and Marvin Jacoby, Norristown, Pa. /
Sperry Rand Corp., a corp. of Del. / A
code generator.
May 31, 1960
2,938,666 / Philip S. Rand, Redding Ridge,
Conn. / Sperry Rand Corp., a corp. of
Del. / A record sensing means.
2,938,667 / Fritz A. Deutsch, East Orange,
N.J. / Monroe Calculating Machine Co.,
Orange, N.J. / A combination card feed
and sensing means.

Following is the index of advertisements. Each item contains:
Name and address of the advertiser / page number where the
a,lvertisement appears / name of agency if any.

Bendix Computer Div., los Angeles 45, Calif. I Page 31 I
Shaw Advertising, Inc.
Berkeley Enterprises, Inc., 815 Washington St., Newtonville
60, Mass. I Page 24 I Electric Boat, A Div. of General Dynamics, Groton, Conn.
I Page 23 I D' Arcy Advertising Co.
General Electric Co., Missile & Space Vehicle Dept., 3198
Chestnut St., Philadelphia 4, Pa. I Page 2 I Deutsch &
Shea, Inc.
International Business Machines Corp., Data Processing
Div., 112 E. POSt Rd., White Plains, N.Y. I Pages 16, 17
I Marsteller, Rickard, Gebhardt & Reed, Inc.

2,938,668 / Byron L. Havens, Closter, and
Charles R. Borders, Alpine, N.J. I
1.B.M. Corp., New York, N.Y. / A
serial-parallel binary-decimal adder.
2,939,001 I William J. Deerhake, Dumont,
and Byron L. Havens, Closter, N.J. /
1.B.M. Corp., New York, N.Y. I A regenerative data storage system.
2,939,081 I Jane H. Dennis, Cambridge,
Mass. / Philco Corp., Philadelphia, Pa .•
I An information storage system.
2,939,082 / Norman Nesenoff, Flushing,
N.Y. I Sperry Rand Corp., Ford Inst. Co.,
Div., a corp. of Del. / An electronic
function generator.
2,939,117 I Edgar A. Brown, Owego, N.Y.
/ I.B.M. Corp., New York, N.Y. I A
magnetic core storage device with flux
controlling auxiliary windings.
2,939,119 / Theodor Einsele, Sindelfingen,
Germany I 1.B.M. Corp., New York.
N.Y. / A core storage matrix.
2,939,120 I Eugeni Estrems, Saint Mande,
and Maurice Papo, Paris, Fr. I I.B.M.
Corp., New York, N.Y. / A data processing machine having a multiple step
program device for controlling the sequence of the operations.
June 7, 1960
2,939,631 / Albert Burstein, Philadelphia,
Pa. and Arnold M. Spielberg, Haddonfield, N. J. I R.C.A., a corp. of Del. I
A data input control system.
2,939,634 I Joseph A. Beek, Jr., Palos
Verdes, and Glenn E. Hagen, Manhattan Beach, Calif. I Alwac International,
Inc., a corp. of Panama I A computer
data control system.
2,939,758 I Loring P. Crosman, Wilton,
Conn. I Sperry Rand Corp., a corp. of
Del. I A magnetic data recording apparatus.
2,940,066 I William F. Steagall, Merchantville, N. ]. / Sperry Rand Corp.,
a corp. of Del. / A bistable device.
2,940,067 I Cecil B. Shelman, Fort Worth,
Tex. I General Dynamics Corp., San
Diego, Calif. I A magnetic circuit for
performing logical functions.
2,940,068 I Rudy C. Stiefel, New York.
N. Y. / Sperry Rand Corp., Ford Inst.
Co. Div., a corp. of Del. I A large
scale memory device.


Minneapolis-Honeywell, Datamatic Div., Wellesley Hills 81,
Mass. I Page 19 I Batten, Barton, Durstine & Osborn
National Cash Register Co., Dayton 9, Ohio I Page 32 I
McCann-Erickson, Inc.
Philco Corp., Government & Industrial Group, Computer
Div., 3900 Welsh Rd., Willow Grove, Pa. I Page 3 I
Maxwell Associates, Inc.
Reeves Soundcraft Corp., Great Pasture Rd., Danbury, Conn.
I Page 5 I The Wexton Co., Inc.
Royal McBee Corp., Data Processing Div., Port Chester,
N.Y. I Page 9 I C J. LaRoche & Co.
Technical Operations, Inc., 3600 M St., N.W., Washington
7, D.C I Page 13 I Dawson Macleod & Stivers
Wheeler-Fairchild, Inc., 610 So. Arroyo Parkway, Pasadena,
Calif. I Page 21 I ~OMPUTERS

and AUTOMATION for November, 1960








Just as the organization chart of any enterprise
provides the framework for dynamic growth
and adaptation, so "organization chart" design
provides the framework for ,unparalleled
expandability in the Bendix G~20 data processing system. This means that your G-20 can economically mat<:h; st'ep by step, your expanding
, scientific or business computational workloads
... without sacrificing system balance.• Key to
, the organization chart efficiency of the Bendix
'G-20 is the ability of the Central Processor to
Hdelegate" routine data handling tasks to control buffers, ,acting'as "'line supervisors." Thus
freed, the Central Processor can make most
efficient use of its high computation speed, its
ability to schedule program priorities and'

... "

direct accessory equipment assignments .• The
ability to employ many control buffers gives
new meaning to expandability in the Bendix
G-20. Up to 70 input/output units can be
directed by each of these "line supervisors."
The Bendix G-20 has a complete line of accessory equipment including keyboard, paper tape,
punched cards, high speed line printers, and,
2 million word magnetic tape units .• Memory
represents another dimension of G.. 20 exp~nd­
ability; ranging from 4,OH6 to 32,768 words.
• Investigate the organization chart design of
the Bendix G-20. See how it provides balanced,
practical expandability ... at a cost that assur~s
unequalled data processing performance, per'~
dollar invested .• For detailed literature write:'

Bendix Computer Division



(F l 0 l , AiR FORCE PHOTO)

National has had many years' experience making significant contributions to the defense effort, including airborne components.

This operation will interest any engineer
o r scientist pos essin g enou gh self-confidence-abilit y and experience-to develop
projects initiall y and carry th em through
to co mpleti on.

N ation al is loo kin g for mi litary-oriented
scientists and enginee rs who hold a B.S.
degree or adva nced degrees. You should
be working in electronic, electro- mechanica l, mechanica l, physics, optics, m athematics, or othe r re lated areas. Preference
will be given to those who have had several yea rs' experience dealin g with prime
con tractors and government age ncies.




As a member of N ational's New Military
Development T ea m-you will be working
initiall y with our Militar y Proposal Group.
As proposals become specific projects,
your responsibility will continue throu gh
th e contractual stage for technical liaison,
fulfi ll ment of contractual obliga tions includin g hardware development, meanwhil e
reta ining sufficient flex ibility to continue
your proposal efforts.

N ation al's new Military Research and Developm ent Program offers you unu sual
latitude in responsibility. It offers you the
chance to participate in military projects






from sta rt to finish. Furthermore, you now
have th e opportunit y to join an operation
still in its form ative stage-yet bac ked by
one of the world 's most successful .••
most reputable corporations.
sending your resume to Mr. T. F . Wade,
Technical Placement Section F5-5 . The
N ational Cash Register Company, D ayton
9, Ohio. All correspondence will be kept
strictly confidenti a l.


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-c041 52.342996, 2008/05/07-21:37:19
Create Date                     : 2015:08:25 19:46:32-08:00
Modify Date                     : 2015:08:25 19:04:30-07:00
Metadata Date                   : 2015:08:25 19:04:30-07:00
Producer                        : Adobe Acrobat 9.0 Paper Capture Plug-in
Format                          : application/pdf
Document ID                     : uuid:725c0426-a208-1e40-80cf-eddcb202b6ba
Instance ID                     : uuid:b22b8259-c135-db47-8243-29716dac6aed
Page Layout                     : SinglePage
Page Mode                       : UseNone
Page Count                      : 48
EXIF Metadata provided by EXIF.tools

Navigation menu