196307
196307 196307
User Manual: 196307
Open the PDF directly: View PDF 
.
Page Count: 56
| Download | |
| Open PDF In Browser | View PDF |
CD
ters
automation
=T~IIII~~I1=
F_fIP
I II F • ,., III fill T I • II
~"III:~==1111i
Give your computers a 3000-mile reach
Put DATA· PHONE service to work for your company and
your business machines can reach anywhere there are
regular telephone lines.
Send any kind of data that can be put on punched
cards or tape-any time of the day or night at regular
telephone rates.
Think how DATA·PHONE service can save you time
and money in shipping vital word and figure data to
and from all your business locations. See one of our Communications Consultants about it. Just call your Bell
Telephone Business Office and ask for him.
Bell Telephone System
..
In case YOtIY eye hasn't the agility of an optical reader,
the title on our cover reads "Standards for Information Processing."
It is lettered by our artist in an imaginative font
that parallels the more familiar MICR numerals.
The standards question is covered from two views in this isStte.
For those already actively following developmellts
in this cOlltroversial field,
Dick Utmall of BEMA presellts a report on currellt progress on page B.
For those tljwt gettillg ill to the know,"
a Primer 011 Stalldards for IP begills on page 13 .
JULY, 1963 Vol. XII, 110. 7
computers and data processors:
the design, applications,
and implications of
information processing systems.
editor and publisher
EDMUND C. BERKELEY
assistant editors
III This Issue
MOSES M. BERLIN
NEIL D. MACDONALD
L. LADD LOVETT
8
contributing editors
13
16
A PRIMER ON STANDARDS AND THE GROUPS PROPOSING THEM
MANAGEMENT, AS A PROBLEM IN PROCESS CONTROL
26
PERT AND THE ROLE OF THE COMPUTERS
STANDARDS FOR INFORMATION PROCESSING: A PROGRESS
REPORT
by R. E. Utman
ANDREW D. BOOTH
NED CHAPIN
JOHN W. CARR, III
ALSTON S. HOUSEHOLDER
PETER KUGEL
advisory committee
T. E. CHEATHAM, JR.
GEORGE E. FORSYTHE
RICHARD W. HAMMING
ALSTON S. HOUSEHOLDER
HERBERT F. MITCHELL, JR.
by Simon Ramo
by Russell D. Archibald
111 Every Issue
across the editor's desk
32
editorial
5
associate publisher
6-7
ANN B. BAKER
art director
N. DORF
cirmlation mallager
VIRGINIA A. NELSON, 815 Washington st.
Newtonville 60, Mass., DEcatur 2-5453
advertising representatives
Chicago 11, COLE, MASON AND DEMING
737 N. Michigan Ave., SU 7-6558
Los Angeles 5, WENTWORTH F. GREEN
300 S. Kenmore Ave., DUnkirk 7-8135
San Francisco 5, A. S. BABCOCK
605 Market St., YUkon 2-3954
Elsewhere, THE PUBLISHER
815 Washington St., DEcatur 2-5453
Newtonville 60, Mass.
Computer Translation from Chinese to English, and Some of Its
Implications
editor's scratchpad
PATRICK J. MCGOVERN
production manager
COMPUTING AND DATA PROCESSING NEWSLETTER
1963 - One Lap Too Many?
readers' and editor's forum
7
7
7
12
21
22
22
48
Who Said That the Computer Doesn't Threaten Man?
Information Bureau on Computer Science
Association for Computing Machinery National Conference, Denver,
Colo., August 27-30
Real-Time Decision Making
"Character" Recognition
ACM Meeting Features "Wi Id West" Tour
Obsolescence of Experienced Engineers and Scientists
Calendar of Coming Events
refere11ce il1formati011
46
49
50
Monthly Computer Census
Books and Other Publications, by Moses M. Berlin
New Patents, by Raymond R. Skolnick
index of 1lOtices
50
52
Advertising Index
Who's Who in the Computer Field
COMPUTERS AND AUTOMATION IS PUBLISHED MONTHLY AT 815 WASHINGTON ST., NEWTONVILLE 60. MASS., BY BERKELEY ENTERPRISES, INC. PRINTED IN U.S.A. SUBSCRIPTION RATES: UNITED STATES,
$15.00 fOR 1 YEAR, $29.00 FOR 2 YEARS, INCLUDING THE JUNE DIRECTORY ISSUE; CANADA, ADD SOc A YEAR FOR POSTAGE; FOREIGN, ADD $1.50 A YEAR FOR POSTAGE. ADDRESS ALL EDITORIAL AND
SUBSCRIPTION MAIL TO BERKELEY ENTERPRISES, INC., 815 WASHINGTON ST., NEWTONVILLE 60, MASS. SEcdND CLASS POSTAGE PAID AT BOSTON, MASS.
POSTMASTER: PLEASE SEND ALL FORMS 3579 TO BERKELEY ENTERPRISES, INC., 815 WASHINGTON ST., NEWTONVILLE 60, MASS. COPYRIGHT. 1963. BY BERKELEY ENTERPRISES, INC. CHANGE OF ADORESS: IF YOUR ADDRESS CHANGES, PLEASE SEND US BOTH YOUR NEW ADDRESS AND YOUR OLD ADDRESS (AS IT APPEARS ON THE MAGAZINE ADDRESS IMPRINT), AND ALLOW THREE WEEKS fOR ..HE
CHANGE TO BE MADE.
COMPUTERS AND AUTOMATION, FOR JULY, 1963
DATAREPRODOCTION
We handle data processing problems from every direction
Look to A-M for EDP input-output compatibility. Low-cost,
practical data processing methods are available to serve a wide
and diverse need . • FOR EXAMPLE: Low-cost Addressograph®
Master Records (Metal and Plastic) or Multilith® Masters
store data for immediate processing. Portable equipment
such as Addressograph Data Recorders collect information accurately at points of origin - reduce clerical
time, eliminate' errors. Fed into the A~M Qptical Code
Reader, data is scanned, punched and verified automatically for direct input to EDP systems. • Process·
ing equipment includes compatible high speed A·M PrinterProcessor systems which free computers for primary on·line
functions and produce 1000 132-character lines per minute.
• A·M duplicators with automatic copy sorters let you
quickly reproduce and distribute computer output data
better and faster. • Whatever direction your data processing takes, consult your nearby A·M representative.
He has the equipment, methods and _systems knowledge to help you find cost-saving short cuts in every
phase of data processing.
VISIT A-M AT THE INTERNATIONAL DATA PROCESSING CONFERENCE-BOOTHS 301 through 305
Addressograph - M ultigraph Corporation
MAIN OFFICE: 1200 BABBITT ROAD, CLEVELAND 17, OHIO • BRANCH OFFICES IN PRINCIPAL CITIES OF THE WORLD
4
COMPUTERS'at?-d AUTOMATION for July, 1963
c&a
EDITORIAL
Computer Translation From Chinese to English,
and Some of Its Implications
/'
Elsewhere in this issue of Computers and Automation
is a report on work done at International Business Machines Corp. under a contract with the Air Force on translation from Chinese to English. The evidence is clear that
fairly good translation by a combination of human being
and computer is being accomplished.
In the first step in the process, persons who do not know
Chinese employ a typewriter-like machine and a geometricrecognition technique, for recognizing first the top and bottom of the Chinese character, then the character as a whole,
and then translating it into an IS-bit code. Second, the
computer by means of a program makes a fairly good translation into English. Third, this translation, if desirable,
can be edited (by a fluent Chinese-English interpreter
called a "post-editor") into satisfactory English, conveying
the same meaning as the original Chinese.
This is a truly remarkable scientific achievement. And
the more so, because the structure of Chinese is markedly
different from the structure of any Indo-European tongue
such as English or Latin, German, br Greek.
Even if the degree of perfection of the translation is
short of a hundred per cent, there are not many human beings who know Chinese, know English, and are able to
translate as fluently and in as many contexts as the computer.
This achievement intimately associates clever people designing the program and the "stupid but powerful beast"
of a C()mputer, ·which can do almost any elementary reasoning operation at speeds of more than 100,000 per second.
This successful association forecasts more possibilities of
importance and interest. Among these possibilities is the
power of an adequately programmed computer to deal not
only with words but also with ideas.
Take for example, recognizing the ideas expressed in
words independently of the words which say them, and
then reacting to the ideas in intellectual ways, such as
COMPUTERS and AUTOMATION for July, 196,3
c1arif)'ill~
thelll, I'elll()\'ill~ illl'OlIsistellcies, or summarizing'
them.
The first step ill this process would he to make the words
listed in the vocabulary with their several Illeanings carry
context labels next. to each meaning. For example "shurne,
trick, deck, suit" have one meaning in the context of
the game of bridge, and entirely different meanings in
other contexts. Context labels for meanings for words are
already necessary in machine translation from one language
to another.
The second step in this prbcess would be to denote the
sequence of meanings asserted by the succession of words
in the sentence, and to identify the subject and the predicate. Thus the computer would attribute to each sentence
the meaning with the highest probability of being the intended meaning, Already single-meaning words and singlemeaning sentences occur in mathematics, logic, and computer programs; it should not be too hard to extend the
number of subjects in which this fruitful condition exists.
The third step is to make specific in terms of computer
programs the meanings of such intellectual operations as
clarifying, making consistent, summarizing; etc. This should
not be too hard-because (1) many human beings learn to
do this in high school English courses, and (2) this process
does not involve the operation especially baffling for computers of looking at a wide field of objects and of scenery,
and selecting what needs to be attended to, such as an approaching car or a traffic 'light.
And so we will be further along the road of making all
the language of thought calculable like mathematics.
5
Mechanical simplicity ... which yields a degree of reliability unattainable
by any other paper tape reader! Simplicity made possible through the
utilization of the revolutionary PMI printed motor direct drive servo.
Movement of the tape through the read head is achieved by merely starting and stopping a printed motor. The brakes, clutches and pinch rollers
that cause big trouble and down time in conventional tape transports
are completely eliminated.
Line by line cycle: movement of tape (A) over read head (B) is controlled by
drive capstan (C)-attached directly to shaft of PlVn printed motor* (D); springloaded rollers (E) hold tape gently against capstan, keeping tape movement in
exact accord with capstan rotation; advance command pulse accelerates motor,
capstan, and tape; as read head detects next sprocket hole, a reverse pulse to
motor halts capstan and tape with next character perfectly aligned in read
head. *U.S. Patents of Printed klotors~ Inc. Pending.
PHONE" WIRE OR WRITE FOR COMPLETE INFORMATION
[0 ~ @~@t ff(((;[(0 ffU'$
CORPORATION
1r&rP'E: rR2EA[D)ElR
G len C
0 V
[O)nVnSn~lX1
e, N.
Y.
EDITOR'S
SCRATCHPAD
Members of "Computers & Automation" staff, like nearly everyone else, like to
boil thoughts off the top of their minds fro~ time to time. Up to the present there
has hardly been a suitable place in the magazine for this distillation. So we have
invented this page for the output.
The purpose of "Scratchpad" is to express comments, notes, observations, and
opinions that relate in some way to the computer field, yet which fail to fit into
a regular category, or which consist largely of personal opinions, and at times
even opinions out on a limb. Elsewhere in this issue appears a Danish proverb worth
repeating here, "Prediction is difficult, especially where the future is concerned.'"
We hope that you, our reader, will find this page interesting and provocative,
and we invite your comments and opinions with a light pen, so that we may have the
benefit of your reactions ••• vo1atile or distilled.
~c:,~
------------------------------------------
1963 - ONE LAP TOO MANY? The computer field may well remember 1963 as the big
year of "drop-outs" in the race to capture a sustaining sector of the computer market. As we turn past the mid-year pole, we see several firms waving the white handkerchief and going back to the pits.
The movement was headed last year by General Mills' brief, but notable, turn
round the track. After about one year of contention, General Mills ran out of financial fuel for its low-horsepower marketing drive on the AD/ECS-37 computer, and
retired from the competition in the Fall of '62.
As the '63 stretch began, Advanced Scientific Instruments, Inc., after almost
two years of independent steering, lost control, and was taken over by Electro-Mechanical Research, Inc., of Florida.
Its ASI-210, and ASI-420 computers are continuing to be marketed through EMR.
By far the biggest change in the race pack this year has been the fusion of the
Computer Division of the Bendix Corp. with Control Data. After almost nine years of
competition, with the familiar G-1S and G-20 capturing over 360 installation sites,
Bendix turned the engine over to that impressive young contender, Control Data, in
mid-March. Now the CDC-1S and CDC-20 are new banners in the computer field.
The following month, the Autonetics Division of North American Aviation waved
the white handkerchief on its RECOMP line of computers. After gearing up with a
spirited marketing campaign in the past two years, Autonetics decided that the rewards for entrants in the scientific and engineering sector of the small computer
market were too meager to justify the effort. They are looking for spectators who'
are willing to help them part with an estimated two RECOMP II computers and eight
RECOMP III computers still in the shop.
Most recently, an escape hatch has been opened by E1-tronics, Inc. when it announced that its ALWAC division was for sale. This "little david" in a competition of
g61iaths, has been holding to a niche in the market for over ten years. The industry had been expecting the oft-rumored ALWAC-IV to replace its vacuum tube ALWACIIIE, but new modef plans have been sidetracked till after potential acquirers of
the division have been sounded out. As of presstime, there was no word of any serious offers for the small, but steadfast competitor.
With a full half-lap more to go this year, there are speculations about who
else may be involved in making a terminal pit stop. Rumors persist that Burroughs
may be acquired by a large automotive corporation, that UNIVAC will definitely be
sold this year to just about any billion dollar corporation you can mention, that
Philco will draw back completely from the commercial market and concentrate on military computers, ·and that Packard Bell may be bleaching its bandanna.
The flagman waving the contendors past the 1964 gate will undoubtedly be in for
some major surprises.
THE SOVIETS TRY A "DECADENT" CAPITALISTIC STEP in their efforts to open a market in Western Europe for their Ural II digital computer and MN-7 analog computer:
"ADVERTISING" !
A recent full page ad in a Soviet export magazine published in the Netherlands
extolled the virtues of the Ural-II as "absolute accuracy", "a split second to do
the most complicated problems", and "convenience and simplicity of servicing".
The Automatic Data Processing Newsletter of the Diebold Group, which first reported the ad, suggests that since the Ural-II is a vacuum tube machine, obsolete
by Western standards, this may be an effort to "dump over-runs or inventory of an
old machine" on a curious West.
Our records show that to date ten Ural-Ills have been installed outside the
Soviet Union: Red China, 1; Czechoslovakia, 2; Ghana, 1; India, 1; Poland, 3;
Roumania, 1; and Syria, l ... namely, entirely to satellite nations and underdeveloped
areas. We shall watch with interest, but with little optimism, the prospects of the
Sovietls competing in the already overcrowded European computer market.
SPEAKING OF ADVERTISING, UNIVAC is letting loose with some high-explosive copy
in its new ad campaign in the computer field, and shelling that legend of the industry ... the IBM sales force. According to UNIVACls ad approach, IBM is clubbed as be~
ing the "most advertised", "most publicized", "most quoted" and "most promoted",
while UNIVAC humbly claims the shibboleth "most advanced". Other ads in the series,
largely aimed at that mythical class top management, brand the IBM salesmen as being
the sharpest talkers, the warmest handshakers, the steadiest backslappers, and the
men you always have with you. UNIVAC merely claims that its less effective salesmen
have a more effective product. We understand that since the campaign began UNIVAC
has added free psychiatric- care for its ~ales. force, as a fringe benefit.
As yet no stirrings have been heard from the munition room below the august
guns of White Plains.
WE GAZE WITH AWE at the wonders of the computer field which, no matter how hard
we try, seem to escape our grasp.
For example we marvel at the computer produced by asexual methods, e.g. 11 • • • data,
converted into electrical pulses, ran into one of the worldls fastest computers, so
complex it was designed and built by another computer". (from "The Death of NikeZeus" by E. Ubell and S. Loory, Saturday Evening Post, June 1, 1963) .
... we marvel at the availability of computers, and their ease of use, which
allows AFIPS to state that "Computers in the United States are available to everyone
-- in government, industry, universities -- even students can have access to a machine".
(from "Fact Sheet On The Electronic Computer Industry", pre-pa,red by AFIPS
Public Information Office, March, 19&3) .
... we marvel at the dedication of people in the computer field, such as that
of Herb Mane loveg of BBDO who recent ly charged co-workers in the media field to "asy.
not what the computer can do for you, but what you can do for the computer".
(from
speech at the Boston Ad ~lub, May 14, 1963).
Conducted by Leichtlicht Schreibfeder
•
c&a
READERS 1 & EDITOR1S FORUM
WHO SAID THAT THE COMPUTER
DOESN'T THREATEN MAN?
:
~~<
~"',.
.,
:
.,
:, I'": ::";:>Y~",:,:i,:',i
The threat of computer-stimulated automation becoming the doom of man has hardly ever been more vividly
illustrated than in the above "photo study." David Mayes
appears to be a victim of an information explosion, carried
on the medium of seven-channel paper tape. Possible "suffocation" occurred during a test run of the Creedomat, a
new high-speed tape-punchingJtape-reading typewriter 'produced by Creed & Company, Limited of Croydon, England.
INFORMATION BUREAU ON
COMPUTER SCIENCE
An information bureau has been established by the
American Federation of Information Processing Societies to
provide people in communication fields and public agencies with accurate, authoritative, unbiased information on
all phases of computer science and technology.
Seventy-five scientists in the computer field have been
selected for their depth of knowledge, perspective, and
ability to communicate, and have pledged to participate.
They will participate by explaining, upon request, the
meaning of new developments, by translating technical
terms, and providing viewpoints on the implications of developments. The spread of rapidly developing computer
technology has made it difficult for editors of technical publications and general media to maintain up-to-date knowlCOMPUTERS and AUTOMATION for July, 1963
edge of new advances; and this Bureau should be of assistance.
The Information Bureau has no present plans for distributing prepared materials but will serve on a "question
and answer" basis and to refer responsible existing literature to meet the specific needs of individual journalists and
public agencies. Among the topics which the Bureau plans
to cover are the following:
Analog Computers
Hardware
.\nificial Intdligcllce
Automata
;\utolllatioll alld COlltrol
;\utolliatioll I'rou'SSl'S
Busilll'sS Data I'rou'ssillg
Codillg
COlllputcr Elelllellts
Computer Programming
Computcr Rdiahility
History and Future of
Computers
IllImanitil's ;\pplications
Ilyhrid COlllplltcrs
III for III a t iOIl Ret ric\'al
I.illl'ar I'rograllllllillg
I.ogic
~Iacltille Trallslatioll
~Iatltematical Applications
and Aspects
Computcr Standards
Computer Systems
Data Acquisition and
Transformation
Data Processing
Design Automation
Digital Computers
Education
Effects of Computers
General Computer Concepts
Medical Applications
Numerical Analysis
Operations Research
Applications
Programming Languages
Scientific and Engineering
Alpplica tions
Simulation
Switching Circuits
Queries for the Information Bureau may be made to
Mrs. Phyllis Huggins, Public Information Director, AFIPS,
P. O. Box 55, Malibu, California (Phone: (213) 394-6412),
who will then direct the questioner to the participating
scientist.
ASSOCIATION FOR COMPUTING MACHINERY
NATIONAL CONFERENCE, DENVER, COLO.
AUGUST 27-30
One hundred thirty computer-oriented scientists and
mathematicians will deliver papers and conduct panel discussions at the 1963 Conference of the 14,000-member Association for Computing Machinery to be held in Denver,
Colorado, August 27-30, 1963. Speakers will include a
number of eminent scientists, industrialists, and teachers
from the United States and abroad.
Latest designs in computer systems and peripheral
equipment will be displayed in the Denver Hilton hotel,
conference headquarters. Exhibitors will inclllde manufacturers of both "hardware" and "software" cquipment, prill-
(Please i IIrn to
Paw~
22)
7
·.. .... : ...-..
•... •• .•
··......:: ..•
··· ... : ••••
• ••
.
•• .. ••
•
·
•
•• •••
•
·
•
••
·
••• ·
•• .
••
.. ••• ••
:
:
:
•
:
.:
:
:•
:
STANDARDS FOR
INFORMATION PROCESSING:
A PROGRESS REPORT
.:
:
:
:
•:•
R.E. Utman
Director of Standards
Data Processing Group
Business Equipment Alanufacturers Association
New York} N. Y .
••
:•
:
:
•
...
•
.:
..
:
..
··.. .:.:.: ..
;:~z=~1. ;~~
1..
!
.-J
~
••
t:·j
.:~1.
.;:,.",1 •
~f~Ii,:::==:,~ ,~~~ f2::J{,!~f~b
r.~
••• : ~;;~f'
.~'.l
t
l:~~Z;Z~;0:~ IIEif!d:1&,,:
fA
·.. ...•.. :...
...
.
• . .
......
• •• .
·• •••.•
•
,,,
8
:
:
:•
:
:
•
'>":l.,.
•
!
COMPUTERS and AUTOMATION for July, 1963
Probably no subject in the computer field has stimulated
as much discussion and controversy in the past year as
((standards." The author, who guides standards work for
the association of computer manufacturers, presents a comprehensive review and projection of current activity in the
development of standards for information processing.
The data processing industry developed through the 1950's without
the discipline of standardization. Although this led to today's stiff competition which has rapidly given the
user more and more for his dollar, it
also entailed many headaches and increased unnecessary costs both for him
and the manufacturer, such as these:
l. Representation of data by different code sets caused files to be
ordered in different ways and
reports to change format when
different computers were used.
2. Incompatibility of mag net i c
tapes has caused manufacturers
to design and produce, and users
to pay for, converters which
could have been eliminated altogether in a more disciplined
framework.
3. The re-programming problem
has made it expensive and timeconsuming to switch from one
computer to a better one with
different logic.
The time was right, then, in 1960
for standardization, particularly of
glossary, character sets and codes, languages, and input-output media, for
the exchange of information between
information processing equipments
and humans. For this reason, the
American Standards Association (ASA)
and the International Organization
for Standardization (ISO) decided in
1960 to form committees nationally
(X3) and internationally (TC 97) on
Computers and Information Processing, whose scopes:
"Standardization of terminology,
problem description, programming
languages, and communication characteristics of computers and information processing devices, equipments
and systems,"
were adopted in organization meetings held in 1960 and 1961. ISO assigned the Secretariat of TC 97 to the
United States and the ASA, which in
turn recognized the Business Equipment Manufacturers Ass 0 cia t ion
(BEMA) as sponsor of X3 and TC 97,
with X3 to develop proposed draft
standards both for the U. S. and the
world.
Nearly simultaneously with the ISO
formation of TC 97, the International
Electro-Technical Commission (IEC)
decided to form TC 53 to propose
standards related to the electrical
characteristics of Information Processing Equipment. The IEC also assigned
the Secretariat for this work to the
USA, and the ASA-US National Committee of the IEC selected the Electronic Industries Association (EIA) as
sponsor, and formed Sectional Committee X6 to handle this work nationally. The scope of X3 was then
clarified so as to include standards of
a logical and physical or non-electrical
nature. Also formed in 1960-61 were
ISO fTC 95 and X4 on Office l\Iachines standards, including the inputoutput media of such. The Ollice i\[a·
chine Group of BEi\L\ was ~clecled as
sponsor; to coordinate work of (,0111,
mon interest :tlllong these bodies, joim
steering committees were finally erea ted
at the national and iuternational
levels.
Optical Character Recognition
In order to break its work down into
pieces of manageable size, X3 has
seven working subcommittees. The
first of these is' X3.1 on Optical Character Recognition (OCR). As an. ex- .
ample of the amount of work going
into the formulation of information
processing standards, X3.1 has met
twenty-two times in its less than three
years of existence. Furthermore, it
has Task Groups on Font Development, Printing, and Applications.
These Task Groups have met approximately as many times as X3.1 itself.
During its last year of work, X3.1 has
made much progress in selecting a
numeric font for which to develop
specifications (principally the decimal
digits and a few special OCR characters) for a proposed American and
ISO standard. There has recently been
much discussion as to whether there
should be a numeric standard followed later by a compatible alphanumeric proposal, or the ultimate alphanumeric font as the first OCR standard.
It was decided in the January-March
1963 meetings to recommend the.' numeric first. X3.1 is trying to complete
COMPUTERS and AUTOMATION for July, 1963
its draft numeriC proposal by the end
of 1963. They have had to move very
carefully because a less than rigorous
subjective proposal would not be acceptable, or, if adopted, would cost
some manufacturers heavily. A .good
standard, on the other hand, would be
a fundamental to the OCR business,
on which it may then move forward.
OCR is a good example of a situatIon
where a balance must be achieved between the choice of optimum parameters for two interrelated sets of
hardware: in this case the imprinter
and the reader. It will probably be
mid-1964 before the subsequent draft
standards are proposed for the letters
and other characters.
Bank Check Specifications
Closely related to the work of X3.1,
Optical Character Recognition, is the
work done hy X:L7 which assumed the
work of the Office ECluipment Manufacturers Committee in conjunction
with the American Bankers Association on MICR (magnetic ink character recognition). In September 1962,
X3.7 presented to X3 two proposed
Amerjcan standards on Print and Bank
Check Specifications for Magnetic
Character Recognition. Since these
had al~eady been adopted by the
American Banking Association and
had become de facto practices, they
were approved unanimously by X3
and submitted to ASA for processing.
A patent application by a manufacturer on the print specifications has
since come to light, however, placing
the two proposals in suspense until
agreements are obtained to place the
techniques in the public domain. Over
six months have gone by and the
agreements have yet to be reached,
but it is hoped now that they will be
reached shortly.
Internationally, the European Computer Manufacturers Ass 0 cia t ion
(ECMA) TC-4 and the European
Bankers Association (EBA) arc carrying out similar work in parallel to
X3.1 ;;lnd X3.7. It is hoped that
ISO fTC 97 fSC 3-Character Recognition-will ·compare the X3.1 work
with the ECMA TC-4 work on a numeric and' subsequent alphanumeric
9
font before proposing any international draft recommendation. X3.1 is
also starting to work more closely with
ECMA TC-4, though participation in
their bi-monthly working parties is
expensive liaison.
In MICR, the French proposal
CMC-7, based on BULL equipment,
has recently been adopted by the EBA
as a European Common Market standard in preference to the proposed
American Standards. This work assures multiple ISO standards, rather
than the immensely desirable goal of
a single international standard.
Character Sets and Data Formats
In the basic technology of Character S(!ts and Data Formats, X3.2
from its inception took as its main
goal the realization of an American
Standard coded character set for the
exchange of information between information processing (IP) systems and
between IP systems and associated
equipment. Using a 19-point set of
crite.ria, the Subcommittee presented
a 128-character set (not all characters
assigned as yet) to X3 in November
1961 in the form of a proposed American standard. The proposal w,as accompanied by s eve r a I appendices
which show the design considerations
and criteria which led to the set and
describe related sets and adaptations.
The proposed set is not the same
as any of the many sets alre,ady in usc,
nor could it be. Compelling reaso~s
dictated various changes to existing
practices. Here are just a few of the
many design considerations which
were used:
Set Size: Need for letters; digits;
programming language characters of
COBOL and ALGOL; data processing
characters like @, $, %; transmission
characters'like null/idle, delete/idle,
"\!\Tho are you," start of message;
carriage COil trol characters like bell,
carriage return, horizontal tab. It was
decided that 128 was a reasonable set
size, rClluiring a 7-level binary coding
scheme.
Collation Sequence: Although no
collation sequence is specified, much
ef\'ort was devoted to requirements of
ordering, a few of which are: digits
should be consecutive, letters should
be ('ons~clltive, "Johns" should collate ahead of "Johnson," control char~lcters should collate together, where
possible the upper case-lower case
arrangemen t of characters on a typewrite.r should correspond to a single
bit difference between characters of
such a pair, etc.
International Considerations: Provision was made for expansion of the
alphabet.
10
Dissent
During much of 1962, dissenting papers were considered, re,sul ting in
minor changes to tl~e proposal and
rebuttals by X3.2. Then on September 11, 1963, X3 voted to submit a
proposed draft standard to ASA. X4,
however, voted against this draft
standard as inappropriate for office
machines use. BEMA, as sponsor of
X3 and X4 recommended delay in
submission to ASA until representations of the code in the principal
media of punched cards, perforated
tape and magne.tic tape could also be
proposed. In January 1963, however;
X3 approved immediate submittal to
ASA, and the code standard was formally submitted to ASA. The ASA
Miscellaneous Standards Board has
since approved it as an American
Standard, and as a final step the ASA
Standards Council is now re,viewing
all aspects of its development and
processing. It is expected to become
a Standard of fundamental significance
shortly.
Future work by X3.2 in 1963 and
later will include expansion and elaboration of the originally adopted set,
resulting in draft proposals for its representation in the principal media. A
perforated tape proposal is expected
to be available to X3 in three months,
magnetic tape in nine months or less,
and punched cards' in mid-1964.
International Character Set
Internationally, the TC 97/Subcommittee 2 corresponding to X3.2,
proposed at its October 1962 Paris
meeting that its 6- and 7-levcl coded
character set (whose 7-level is almost
identical with that of X3.2, if you
in terpret $ as a special case of the
general currency symbol) be distributed as a draft proposal for an international standard. It also plans to
work on the representation of the
coded set in the various physiGtl
media, beginning at a Fall 1963 meeting.
Digital Data Transmission
In digital data transmission standards work, there is considerable overlap between X3.3 and EIA's TR-27;3.
To aV9id wasteful duplication, the
two groups Inve been holding joint
meetings. X3.3 has created six Task
Groups to reflect its current work and
objectives: Liaison, Glossary, Data
Transmission
Formats, End-to-End
Control Characteristics, System Performance, Digital Data Transmission
Speeds. In June of 1962, it presented
X3 with a proposed American Standard on Signaling Speeds for Data
Transmission, which cleared promptly
through X3, BEMA and ASA and, on
August 8, 1962, became the first Amer-
ican Standard in the Information
Processing field. 1 t is now being circulated internationally by ISO /TC
97/SC 6 for vote as a draft Future
Guides on Signaling Speeds of Data
Transmission, to be available as such
to the ISO-IEC mailing list, including
the International Telegraph and Telephone Consultative Com mit tee
(CCITT), which claims pre-eminence
in this special field of international
standardization.
Programming L.anguages
As an important area for standardization, programming languages
(X3.4) are certainly among the most
difficul t.
Reflecting the common
User desire for such standardization,
three languages, which are. to some
extent machine independent, have
realized wide national and to some
ext en t international recognition:
FORTRAN, COBOL and ALGOL.
FOR TRAN, originally de.signed for
one machine, became so use£.ul that it
has been implemented for about 50
machine type's. Needless to say, partly
because of different machine parameters (memory size, numbers of tape
transports, etc.), partly because the
original standardizing body SHARE
was set up for a single machine family, there is tremendous variation in
these various FOR TRANS. Only with
FORTRAN is the body ot American
usage experience significant. COBOL,
and to a less extent ALGOL, are as
yet too young. N everthc.1ess, there
are also about 50 compilers for each
of them, and ALGOL is achieving
wide usage as a standard language for
the publication of algorithms. Much
experience for these compilers and
languages will become available during
1963 and 1964.
Task Croups
X3.4 is addressing itself, with great
respect for the difficulty of the task,
to the job of standardizing Programming Languages. I t has several Task
Groups. The first, X3.4.l, has the
function of specifying how to specify
a language. A paper by its Chairman,
Saul Gorn, illustrates the variety of
approaches that could be used. Recently, progress has been made, toward solution of the problem of recognizing ambiguities and inconsistencies in languages. This is nOit a trivial
problem.
The second, X3.4.2, is
charged with reviewing draft standards proposed by other X3.4 subcommittees for completeness and acceptability, and also for establishing USA
positions in regard to ALGOL. X3.4
recently reaffirmed its desire that the
USA support ALGOL first as a potential international standard, and
when and if the need arises to propose
COMPUTERS and AUTOMATION for July, 1963
AMERICAN
CHART OF PROGRESS
X3
SC-
Date of
Dates of Approval By
X3 Working Group
ASA
Abbrev.
Complete
Standards
Title
Devel.
Draft
Sponsor Board ASA
X3
pASSC-
Proposed American Standard
X3.l OCR Numeric Font, Print Specifications
62-63
X3.l OCR Alphanumeric Font, Print Specifications
62-64
X3.2 Code for Information Interchange
61-62
X3.2 Addenda to the pASCII-X3.2
Extension Rules (Escape)
62-63
Controls, Definition of
X3.2
63
X3.2
Subsets, Related
63
Unassigned Area, Use of
X3.2
63
(X3.3)
Controls for Synchronous Data Transmission
63
X3.2 Input-Output Media-Perforated Tape Code
X3.2
62-63
X3.2
Physical - Dimensional Spec's (1" )
60-63
(11/16" )60-63
X3.2
X3.2 Magnetic Tape Code
62-63
X3.2
Physical-Dimensional Spec's (112")
60-63
X3.2 Punched Card Physical-Dimensional Spec's (80 & 90 Col.)
61-63
X3.2 Edge Punched Card Physical-Dimensional Spec's
63-64
X3.2 Punched Card Code, ASCII Direct Representation
62-64
X3.2
, ASCII Hollerith Extension
62-64
X3.2 *NMTC Perforated Tape' Code
60-64
NMTC
M.agnetic
Tape
Code
X3.2
63-65
X3.7 MCR (Numeric Font) Print Specifications
60-62
X3.7 MCR Bank Check Specifications
60-62
X3.7 MCR Error Correction Techniques
63-64
X3.7 MCR Bank Check Serial Number
63-64
X3.3 Signalling Speeds for Data Transmission
61-62
X3.3 Data Transmission Formats
62-64
X3.3 Bit Sequencing of the ASCII {7-leveU
62-63
X3.3
Related Subsets
62-64
X3.3 Graphic Presentations of Error Statistics of Rates
62-63
X3.4 FORTRAN II
62-64
X3.4 FORTRAN IV
62-64
X3.4 ALGOL
64
X3.4 COBOL
63-65
X3.4 APT
63-65
X3.5 Glossary for Information Processing
63-64
X3.6 F,lowchart Symbols for Information Processing
61-63
*NMTC
= Numerical
TC97/SC
TC97/SC 2
TC97/SC
B
TC97/SC 4
12/63
4/64
5/64
7/64
8/64
NFOCR
11161
1163
3/63
4/63
5/63
CIl
6/63
6/63
8/63
8/63
8/63
8/63
6/63
10/63
10/63
5/63
6/63
6/63
10/63
10/63
6/63
6/63
8/63
8/63
12/63
12/63
8/63
10/63
10/63
10/63
7/62
7/62
8/62
8/62
9/62
9/62
9/62
9/62
3/62
4/62
6/62
7/62
7/62
5/(>3
11/(,)
LO/(,)
LLIC,)
ClIRev-E
CIIRev-C
CIIRev-S
CIIRev-U
CIIRev-CS
10/63
8/62
8/62
10/63
10/63
216',
3/63
PTC
PTS-l"
PTS-ll/16"
MTC
MTS-1I2"
PCS-80/90
EPCS
PCC-D
PCC-H
NMTCCPT
NMTCCMT
PSMCR
BCSMCR
PSMCRRev-E
BCSMCRRev_S
8/62 SSDT
\DTF
ASCIIRev-BS
ASCIIRev-SB
GPESRDT
FORTRAN II
FORTRAN IV
ALGOL
COBOL
APT
GIl'
LU(,) FSIl'
Machine Tool Control
Date
of
ISO/IEC
INTERNATIONAL
CHART OF PROGRESS
,ISO/IEC
12/63
Late '64
11/61
Draft Recommendations
WG
Complete
Devel. Draft
Dates
Approved By
ISO/IEC
ISOIIEC
Draft fll Draft if2 Draft
General
WG
Proposal Proposal Rcmnd'n Approval
SC-
Glossary
60-64
4/64
Codes
Char. Recog.
TC97/SC 5
1-0 Media
Prog. Lang's
TC97/SC 6
TC95/SC 8
Data Trans.
1-0 Media
Multi-lingual Vocabulary
6 and 7 Bit Coded Character Sets for
Information Interchange
OCR Numeric Font
OCR Alphanumeric Font
MCR Numeric Font (E 13 B)
MCR Alphanumeric Font (CMC 7)
Punched Cards Physical-Dimensional
ALGOL
FORTRAN
COBOL
Guide for Future Signalling Speeds for D. T.
Perforated Tape,' Physical-Dimensional
62-65
62-67
62-65
62-66
62-64
62-64
63-65
63-66
62-63
63-64
Magnetic Tape, Physical-Dimensional
63-65
61-62
1164
10/62
10/62
5/64
10/62 10/62 6163
10/62 10/62 6/63
IFIP/ICC
Vocabulary
CCSII
OCRNF
OCRAF
MCRNF
MCRAF
PCS
ALGOL
FORTRAN
COBOL
SS Guide
PTS
IECI
TC53/SC D 1-0 Media
the ISO standard as an American
Standard ALGOL.
X3.4.3 FORTRAN was created in
August 1962 and is hard at work defining two levels of FORTRAN as a
draft American Standard. Membership in X3.4.3 and its task groups is
open to an Implementer and a User
Group representative associated with
every USA compiler. Associate membership is also available to representa-
tives of non-USA FORTRAN compilers.
X3.4.4, COBOL, was formed in
January 1963 to prepare a COBOL
proposal for processing as an American standard. The COBOL Maintenance Committee of CODASYL will
remain the maintenance and defining body for COBOL, whereas X3.4.4
will create test problems and prepare
the specifications of the draft stand-
COMPUTERS and AUTOMATION for July, 1963
MrS
ard. Close coordination with CODASYL is assured through a joint steering committee, and internationally
with ECMA/TC-6-COBOL through
periodic meetings.
X3.4.5 is the American secretariat
and delegation to TC 97/SC 5, Programming Languages. Preparations
are nearing fOJllpletiol1 for the fourth
meeting at the ISO /TC 97 level, to
be held ill June ill Berlin.
It
X3.4.6 works on Glossary concepts
in the Prog-ramming- Lang-uage field.
Internatianally, TV 97/SC 5 has
prepared a Prog-ramming- Language
Survey which is now being published
in professianal jaurnals, and has invited IFIP /WG 2.1 (a Working Group
.of the Internatianal Federatian .of Infarmation Processing Sacieties) and
X3.4 ta pravide draft prapasals far
ALGOL, FORTRAN and COBOL
for consideration as inte.rnational
draft recommendatians. It is about
to conduct a depth survey on these
three languages.
Glossary
In glassary work, under the new
chairmanship of Martin Weik .of the
Department of Defense, X3.5 is working on plans to produce a proposed
American standard glossary for informatian processing by the end .of 1963.
They are merging several major domestic and internatianal vacabularies
(ACM, IFIP, IRE, etc.) with the, specialized glossary requirements of the
ather X3 subcommittees into an acceptable whole. These input definitians are punched into cards at
BEMA, and the compilation process
is mechanized, with merged listings
resulting for X3.5 analysis. Their final
recommendation, tfrm-by-term, is then
circulated to each X3 subcommittee
for criticism and acceptance. For interim use by X3, the Government Interagency ADP Cauncil Glossary has
recently been approved by X3.5.
Interna'lionally, TC 97/SC 1 has a
different multi-lingual approach and
objective. It recognizes the wark .of
the IFIP-ICC glossary develapment
effart, which proceeds by defining
concepts to whi"ch specialists in each
language (English, French, German,
Italian) attach words. This work is
rapidly gathering momentum and it
appears that by May 1964 a fairly
goad international Informatian Processing vocabulary will be available.
It w,illcontain many coined words or
words from one language adopted in
another. X3.5 has recently formed a
task group (X3.5.2) to coordinate USA
participation in this IFIP /ICC effort.
Internationally, at the TC 97 Octaber Paris meeting, the f.ormatian of
a counterpart ISO/TC 97/WG GProblem Definitian and Analysis
group was approved. It will also start
by considering Flaw Chart S.ymbol
draft propasals from the Netherlands,
IFIP, ECMA and the USA. No meetings have been scheduled yet.
Input-Output Media
At the internatianal level 1-0 media
are the concern .of ISO/TC 39-Machine Taals, TC 95-0ffice Machines,
TC 97-Camputers and Information
Pracessing, and .of IEC/TC 44-Electric
Equipment of Machine Tools and TC
53-Camputers and Informatian Processing. In the United States X3, X4
and X6 have been equally concerned
in subcommittees X3.2, X4-A4, and
TR-27.6. I t is natural, therefore, that
companies, countries and ather interested graups wha are finding it hard
to man so many committees should
have pressed for some simplification
.of the administrative procedures. The
Joint Steering Cammittees at the national and internatianal levels, created to solve such problems, have
recommended formatian of joint
working groups to develap 1-0 media
standards. These did not prove productive domestically. X3 recently recommended consolidation of X3, X4
and X6 1-0 media respon·sibilities
within X3, with the concurrence of
X6, and canditioned agreement of X4.
ASA then wen t further and assigned
all 1-0 standardization responsibility
to X3, including physical and electrical characteristics, equipment, and
code representation. X3 in turn decided on May 2 to put all 1-0 and
Codes responsibility into a reorganized
X3.2, which will also be responsible
for all USA participation in the numerous internatianal working groups
and subcommittees mentioned pre,viously. It is hoped that this remarkable and rare oppol'tunity for consolidation and simplification just
experienced nationally can next be
extended to the international level.
The new X3.2 is immediately confronted with processing draft American and international proposals for
physical-dimensional standards in perforated tape, magnetic tape, and
punched cards, and for numerically
controlled machine tool punched tape
code, resultant from four previous
years of EIA and joint working group
development. This processing and
evalua'tion work will be undertaken
as soon as qualified engineering becomes available within the membership of the new X3.2. Participation
in several
international working
group 'meetings later this year must
also be immediately organized, with
delegates and USA positions or proposals provided.
Process Control
The scope of X3 was expanded
£ormall y in January 1963 to include
responsibility for standards in the field
of process control, including numerical machine tool con tral. X3 membership is in the process of expansion
to include six new members from this
industry and its users and general interests. X3.2 is expanding its membership to enable processing code and
media requirements in process control systems. X3.4 has already begun
a study project on the potential of
APT (Autamatically Positioned Tool)
as a standard programming language
for machine tool applications. This
new X3 responsibility is expected to
influence all subcommittees, and eventually require new groups for specialized ,needs.
REAL-TIME DECISION MAKING
1'~4W?l
~
o
o
o
Problem Description and Analysis
X3.6 on Problem Description and
Analysis has devoted the last two and
a half years and twenty meetings to
the dc.velopmen t of a propased American standard flow chart symbolism for
information processing, which is now
about to be circulated to X3 for official vote on acceptability. The future
of this subcommittee is as broad as its
field of systems analysis and description. They are currently evaluating
their fidd for the next most important need in standards.
12
"Before I read this answer, may I say I sincerely hope all of you
gentlemen have bus fare . . ."
COMPUTERS and AUTOMATION for July, 1963
A PRIMER ON STANDARDS
AND THE GROUPS
PROPOSING THEM
Although the call for "standards" has
been reaching increasing volume during the last several r11;onths, many people are still unaware of what standards
are, who proposes thern, and how they
are reviewed, accepted, or rejected. Below is an outline of the groups active
in standards for information processing in this country, the subjects they
are wOl-hing on, and their membership.
•
The American Standards Association
(ASA) is a privately supported organization acting as the national clearinghollse
and coordinating agency for volun tary
standards in the United States.
AS.-\ is a federation of 138 trade association and professional societies. Over
2,000 companies are sllstaining members.
AS'-\'s main functions are:
1. To provide systematic means for the
development of American Standards
2. To promote the devclopmen t and lise
of national standardization in the
United States
3. To approve standards as .-\merican
Standards provided they are accepted
by a consensus of all national groups
substantially concerned with their
scope and provisions
4. To coordinate standardization activities in the United States
5. To serve as a clearinghouse for information on American and foreign
standards
6. To represent American interests in
international standards work
More than 2,000 American Standards
have been developed and approved under
:\S.-\ procedures. These standards apply
111
the fields of engineering, industry,
safety, and consumer goods.
HISTORY
. In 1918, five leading American engineering societies decided to form a national organization that could coordinate
the development of national standards,
founding the ".-\merican Engineering
Standards Committee"-the forerunner of
the .\merican Standards .-\ssociation.
Three departments of the federal government, Commerce, "'ar, and :'-Javy, joined
the organization as founding members.
In 1928 the American Engineering
Standards Committee was. reorganized
and renamed the .-\merican Standards Association (ASA).
In 1948 .-\SA was incorporated under
the laws of the State of :'-Jew York. A
Board of Directors is responsible for policy, administration and financial matters
.\. Standards Council, representing ali
member-bodies, supervises all technical
activities and determines the over-all
standards policy.
About 400 standards projects are currently active. Approximately 10,000 engineering, government officials, and representatives of various national groups are
participating in these projects.
A~1ERICAN
STANDARDS
An American Standard is a voluntary
national standard, arrived at by common
consent, and available for voluntary use.
It is a standard approved by the American Standards Association. Because ASA's
procedures are based upon the principle
of consensus from the initiation of work
to its conclusion, an American Standard
is accepted as the one national authoritative standard in its immediate field of
application.
An American Standard can be developed by a committee set up under the
auspices of the American Standards Association.
Also, tra(\c associations and professional
societies which have developed their OWII
standards call suhmit these to AS.\ for
approval as American Stalldards.
Approval of a standard as "American
Standard"-no matter how or where developed-is given ollly if the standard is
supported by a consensus of all national
groups substantially concerned with its
scope and provisions.
I~TERNATIONAL
STA:'-JDARDIZATION
The American Standards Association
holds the U. S. membership in two international standardization bodies-International Organization for Standardization
(ISO), International Electrotechnical
Commission (IEC)-and the Pan American Standards Committee (P ASC). In
this wa y, ASA provides the channel
through which American interests can
participate in global and hemispheric
standards projects. ASA also works with
the British Standards Institution and the
Canadian Standards Association through
the "ABC" (American-British-Canadian)
Conference on Unification of Engineering
Standards.
The ASA library is a source of information on thousands of foreign standards.
The Association supports U. S. foreign
trade by promoting a knowledge of
American Standarcls abroad.
HOW AMERICAN STANDARDS
ARE ACHIEVED
American Standards come into existence through three basic methods.
1. Section Committee Method. A committee, composed of representatives accredited for the purpose by all groups and
organizations substantially concerned with
the scope of the standards project and organized under the rules of ASA for such
committees, formulates the standard. The
special utility of the method consists in
COMPUTERS and AUTOMATION for July, 1963
the provision, in advance, of such representation that a consensus will be as-'
sured and self-evident when the members
have approved their completed assignment.
2. Existing Standards Method. Under
this method, an existing standard may be
submitted for approval by any responsible
body, and may be approved by ASA provided:
(a) It is shown that the standard is
supported by the necessary consensus of those substantially concerned with it
(b) It does not conflict with any other
American Standard
Approximately one third of the standards approved by ASA have been considered under this method.
3. General Acceptance Method. Comparatively simple standards that do not
require prolonged committee work can
he approved al a (;cneral Conference of
all groups vitally ('ol\(l'1'IInl with the scope
of the standard. The wllferellce action
is finalil.cd Ihrough Icllcr hallot action to
esta hi ish I he ex iSlellce of t he consensus
l"C(luired for approval hy ASA.
ASA SECTIO:'-JAL
CO~1MITTEE METHOD
The Sectional Committee Method, one
of the methods recognized by ASA Procedure as meeting the basic requirements
of the Association, consists in the formation, at the beginning of a project, of a
committee to develop one or more standards under an assigned scope. The committee is composed of representatives accredited for the purpose by the various
organized groups concerned with the project and, when desirable, companies and
specially qualified individual~ as general
interests. Membership in a sectional committee may also be in the name of the
organization as such, no individual being
designated as representative or alternate.
ASA is prohibited by its constitution
from formulating standards. It is not a
technical society engaged in standardization work. It therefore cannot own any
committees that tormulate standards.
Thus sectional committees can only be
considered as belonging to the group of
organizations having represelltalion on
the commillec and which havc agreed to
cooperate, under AS.\ procedure, ill the
developmellt of standards IIH'Y all desire.
ASA SECTIO~.\1.
SI'O~SOR
CO~DIITTEE
One or more organil.ations principally
concerned with the work assigned to a
sectional committee may be designated to
give administrative support and direction
to the committee. The sponsor organizat ion is responsible for the administration
and direction of the standards project. It
organizes the s~ctional committee with tht'
13
advice and assistance of ASA; it ensures
that the work is carried out continuously
and effectively; it provides the necessary
administrative services; and keeps ASA
informed on the progress of the work.
A project may have more than one sponsor.
X3 SECTIONAL COMMITTEE
X3
TITLE-CO~IPUTERS
AND
INFORMATION PROCESSING
X3 SCOPEStandardization of the terminology,
problem description, programming languages, comm unica tion characteristics (I),
and physical (non-electrical) characteristics (2), of computers and data processing devices, equipments and systems (3).
Note 1. Includes standardization of symbology, coded character sets and representations, input/output media and formats, character recognition.
Note 2. Includes standardization of the
logical and physical characteristics.
Note 3. Includes process control systems and others as developed.
X3 MEMBERSHIP
Membership is by national association,
society or organization, divided equally
into three groupings by interest:
General Interest Organizations,
Man ufacturer Associations,
User Organizations.
The following organizations are regular
voting members of X3:
GENERAL INTEREST GROUP
(10 Votes)
American Institute of Electrical Engineers
Association for Computing :\lachinery
Association of Consulting ~Ianagement
Engineers
American Management Association
Data Processing Management Association
Department of Defense
Electronic Industries Association
Engineers Joint Council
Institute of Radio Engineers
Telephone Group
MANUFACTURER GROUP (10 Votes)
*'Business Equipment Manufacturers
Association
USER GROUP (9 Votes)
American Bankers Association
American Gas Association & Edison
Electric Institute
American Petroleum Institute
Air Transport Association
General Services Administration
Insurance Accoun ting & Statistical Association
Joint Users Group
Life Office Management Association
;\1ational Retail Merchants Association
*.\1 present one association, BEMA,
represents the interests of manufacturers
in X:l. The ten votes of BEMA are held
hy ten member companies, selected anIlIlally, and including:
Burroughs Corporation
International Business Machines Corp.
Minneapolis-Honeywell Regulator Co.
Monroe Calculating Machine Co.
National Cash Register
Pitney-Bowes, Inc.
Radio Corporation of America
Remington Rand Division of Sperry
Rand
Royal McBee Corporation
Standard Register Company
14
X3 OFFICERS
Chairman, C ..-\. Phillips, BEMA
Secretary, R. E. Utman, BEMA
X3 SUBCOMMITTEES
X3 has organized into seven subcommittees, responsihle for the technical work
of considering and providing for the initial standardization requirements of information processing:
X3.l-0ptical Character Recognition
X3.2-Coded Character Sets and Data
Formats
X3.3-Data Transmission
X3.4-Common Problem-Oriented Programming Languages
X3.5-Terminology . and Glossary
X3.6-Problem Description and Analysis
X3.7-~lagnetic Character Recognition
Each subcommittee has defined its range
of technical responsibilities and activities
in the form of scopes and programs of
work approved by X3. In most subcommitees further subdivision into working
groups has been necessary to the accomplishment of technical purposes (as
shown below). In aU subcommittees and
working groups, members function as individuals rather than official representatives of organizations, although a reasonable balance of interests is desired.
X3.l Title: Optical Character Recognition
Scope: Printed input and output to
data processing systems for interchange of
information between data processing and
associated equipment. Humanly legible
prin ted oharacter sets, e.g., character recognition.
Chairman: J. J. Eachus, BEMA/Minneapolis-Honeywell
Secretary: Terrance Trickett, American
Machine & Foundry Co.
USA Representative: E. Daniel Spina,
BEMA/IBM
\Vorking Groups:
X3.1.1-Font Development
X3.1.2-Printing Capabilities
X3.1.3-Format and Applications
X3.2 Title: Coded Character Sets and
Data Formats
Scope: Recommendation of standards
for coded character sets and recording
formats (Note I) for the interchange of
information between information processing systems and between systems and
associated equi1pment (Note 2). Such recommended standards are to include:
a. The machine sensible coded character set.
b. Such standard formats and format
indicators as are required to define
data fields, data records, program instructions and tche like.
Note 1. Includes responsibility for the
(logical) representation of codes in the
several media, taking into account the
need for error ohecking. It is recognized
that media present physical limitations,
and that close working liaison with a"ther
groups (I/O Media) will be necessary.
Note 2. Includes process control systems.
Chairman: L. L. Griffin, X3/DOD
Secretary: Rotates
USA Representative: John Auwaerter,
Teletype Corporation
'Working Groups:
X3.2.l-Code Representation/~Iagnetic
Tape
X3.2.2-C 0 d e Representation/Perforated Tape
X3.2.3-Code Representation/Punched
Cards
. X3.2.4-Character Assignments/UnasSigned area (pASCII)
X3.2.5-Coded Character Subsets and
A p P lie d Character Sets (related to
pASCH)
X3.2.6-Definition of Control Character Functions (pASCII)
X3.3 Title: Data Transmission
Scope: Determine and define the operational characteristics governing the perfOrn~a!lCe of digital da.ta generating and
receIvmg systems combll1ed with communication systems.
Chairman: A. Stillman, BEMA/RCA
Secretary: E. H. Lohse, BEMA/Burroughs
USA Representative: F. War den,
BEMA/IBM
Working Groups:
X3.3.l-Liaison (EIA & CCITT)
X3.3.2-Glossary
X3.3.3-Description of Equipments
X3.3.4-Establishment of Interfaces
X3.3.5-System Performance
X3.4 Title: Common Programming Languages
Scope: Standardization of common program languages of broad utility through
standard methods of specification with
prOVISIOn for revision, expansion, and improvement, and for definition and approval of test problems.
Chairman: R. F_ Clippinger, BEMA/
M-H
Secretary: K. Speierman, BEMA/Burroughs
USA Representative: H. Bromberg,
BEMA/RCA
Working Groups:
X3.4.I-Language Theory
X3.4.2-L an g u age Specifications &
ALGOL·
X3.4.3-FOR TRAN
X3.4.4-Processor S p e c i fi cat ion s &
COBOL
X3.4.5-USA in ISO/TC 97/SC 5
X3.4.6-Programming Glossary
X3.5 Title: Terminology & Glossary
Scope:
(a) To coordinate and advise the other
subcommittees of ASA X3 in the
establishment of definitions required for their proposed standards.
(b) To recommend to X3 a general
glossary of data processing terms
(for the professional community).
Chairman: Martin H. Weik, X3/DOD
Secretary: J. F. Traub,Bell Laboratories
USA Representative: J. F. Traub, Bell
Laboratories
'Working Groups:
X3.5.l-Definition :\laintenance
X3.5.2-Glossary Advisory Committee
X3.6 Title: Problem Description &
Analysis
Scope: Information Processing Problem
Description and Analysis Standards to
provide a systematic means of studying
information processing problems, documenting, and preparing the required information for analysis.
Chairman: R. W. Green, BEMA/Standani Register
Secretary: J. W. Dresch, BEMA/
UNIVAC
USA Rcpresen tative: J. Pfaff, BEMA/
Burroughs
\Vorking Groups:
X3.6.l-~lethodology
X3.6.2-lnput/Output
X3.6.3-Data Transformation
X3.6.4-Glossary & Terminology
X3.6.5-Flow Chart Symbols
~3.7 Title: Magnetic Character RecognitIOn
Scope:
1. Development of standards for MICR
(present and future).
2. Resolution of problems which may
arise in industry and the market place
which involve the manufacturers and
printers.
Chairman: T. M. Butler, X3/Burroughs
COMPUTERS and AUTOMATION for July, 1963
..
•
Secretary: G. A. Baird, BEMA/Burroughs
USA Representative: E. Daniel Spina,
PRESTOSEAL ANNOUNCES THE
BE~L\/IBM
Working Group:
X3.7.l-Technical
X3 ADVISORY CO~IMITTEE
X3-L\C Title: X3-International Advisory
Committee
Scope:
I. To ensure that there is active participation by the U. S. in the working
groups or subcommittees of ISO/TC 97.
2. To ensure that where the U. S. has
the responsibility for the secretariat of
working groups or subcommittees, the
appropriate subcommittee of X3 is actively fulfilling its responsibilities.
3. 'Vhere there is not an exact correspondence between the scopes of ISO /TC
97 working groups or subcommittees and
X3 subcommittees, to propose methods of
ensuring active participation in, and representation on, the ISO /TC 97 working
groups or subcommittees.
4. By coordination of the activities of
the U. S. representatives on the ISO/TC
97 working groups and/or subcommittees
to formulate for approval by X3 and the
DPG Engineering Committee U. S. policy and to prepare, or have prepared, the
U. S. contributions or recommendations
to the Plenary conferences of ISO/TC 97.
5. To propose the U. S. delegation to
such Plenary conferences.
6. To participate in liaison activities
with other international organizations as
requested by the secretariat of ISO/TC
97.
7. To recommend for approval by X3
and the DPG Engineering Committee all
steps that should be taken to adequately
fulfill, the U. S. role in international
standardization activities.
Chairman: B. W. Pollard, BEMA/Burroughs
Secretary: R. E. Utman, BEMA
l\fembership: USA Chairmen of ISO/TC
97 Subcommittees
SC 3-R . .J. Mindlin
SC 5-R. F. Clippinger
SC 6-A. H. Stillman
USA Representatives
X3 Chairman (ex officio)
X3 I~ JOINT ACTIVITY ON STANDARDS OF COMl\10N INTEREST
NJSC Title: National Joint Steering Committee
Scope: Provide guidance to joint standardization work of common interest
among the Sectional Committees, X3, X4
and X6. Resolve jurisdictional problems.
Chairman: Rotates among chairmen of
X3, X4 and X6
Secretary: Rotates among secretaries of
X3, X4 and X6
Membership: Chairman, Sec ret a r y ,
Technical Advisor of X3, X4 and X6,
ASA Miscellaneous Standards Board Secretary
NJWG/PC Title: National Joint Working Group/Punched Cards
Chairman: Charles Whitaker, NCR
Membership: EIA/TR 27.6 and AS:\
X3.2
NJWG/PT Title: National Joint Working Group/Punched Tape
Chairman: F. W. Williams, IBM
Membership: EIA/TR 27.6 and AS.-\
X3.2
NJWG/MT Title: National Joint Working Group/Magnetic Tape
Chairman: Harry Hayman, NASA
Membership: EIA/TR 27.6 and ASA
X3.2
Joint Activity on Data Transmission
Joint meetings of ASA X3.3 and
DELUXE EDITING CONSOLE
FOR PAPER AND MAGNETIC TAPE
Variable Speed - from 1 in. per sec. to 200 in. per sec.
MODEL
BOOTH
NO.
220
429
SHOWN FOR THE FIRST TIME
TO THE QATA PROCESSJNG INDUSTRY
Af THE J 963 INTERNATIONAl. DATA PROCESSING CONfERENCE
BUSINESS EXPOSITION • COBO HALL • DETROIT • JUNE 25 TO 28
-
PRESTOSEAL MANUFACTURING CORP.
37-12 lOath STREET, CORONA 68. N. Y.
Illinois 7-5566
EIA/TR 27.3 are held for purposes of
standards in data transmission.
Chairman: Alternates between
ASA X3.3 Chairman-A. H. Stillman,
BE~IA/RCA
EL\/TR 27.3 Chairman-j. L.
Wheeler, Xerox
BSRB Title: BE~IA Standards Review
Board
Chairman: Either Director O~IG or Di·
rector DPG, as indicated
Secretary: R. E. Ulman, BE:\IA
~Iembers: Engineering COlllmittees of
OMG and DPG
BE~L\/DPG Title:
BEMA/Data Processing Group
Chairman (1903): R. G. Chollar, ;o..:CR
Vice Chairman (1963): M. G. Mengel,
Burroughs
Director: C. A. Phillips, BEMA
Director of Standards: R. E. Utman,
BEMA
Members:
Addressograph-Multigraph Corporation
Burroughs Corporation
Farrington Electronics, Inc.
Friden, Inc.
International Business Machines Corporation
l\Iinneapolis-Honeywell Regulator Company EDPD
~'1onroe Calculating Machine Company
;\IoOI'e Corporation Ltd.
National Cash Register Company
Pitney- Bowes, Incorporated
Radio Corporation of America
Remington Rand Division of Sperry
Rand Corporation
Royal McBee Corporation
Smith Corona Marchant, Incorporated
Standard Register Company
UARCO, Incorporated
COMPUTERS apd AUTOMATION for July, 1963
Underwood Corporation
Xerox Corporation
BEMA/DPG-EC Title: BEMA/DPG-Engineering Committee
Chairman: C. A. Phillips, BEMA
Secretary: R. E. Utman, BEMA
Members: Representatives of Member
Companies
Scope: Acts for sponsor as administrative ;l\lthority for ASA X3.
We will be happy to send
a complimentary copy of
COMPUTERS & AUTOMAYlON
in your name
to a friend who might find the
information in it stimulating
and useful to him.
Just send his name and address to: V. E.. Nelson, Computers & Automation, 815 Washington Street, Newtonville 60,
Mass.
He'll appreciate
your thoughtfulness!
15
MANAGEMENT, AS A PROBLEM IN PROCESS CONTROL
Simon Ramo
Vice Chai'rman of the Board
Thompson Ramo Wooldridge Inc.
Canoga Park~ Calif.
A leading exponent of the blending of the
versatility of the human mind with the information-processing capacity of electronics
addresses himself to a key question: "What
will happen to management In an automated information world?"
We can expect that, in the future, the world's operations
will become dependent upon, and will be designed around,
a mass extension of the human intellect by electronics. It
may take a decade or more, but eventually information will
be stored efficiently and retrieved virtually instantaneously.
It will be processed in accordance with specific, recorded
rules. An orderly arrangement of the facts of the past will
be compared to the happenings of the present to predict a
substantial portion of the future with useful accuracy.
Electronic systems will solve what today appears to be impending chaos in the paperwork necessary for the world's
production, communication, transportation, professional,
financial, and government operations.
Man-Machine Partnership
However, it is not correct to conceive of the world of the
future as an automatic world in which intellec.tronics equipment has replaced all human brains and senses. Rather, as
the word "intellectronics" implies, the human "intellect"
will be extended by "electronics" to create a new manmachine partnership. Together, we and the machines, in
a sensible co-operative pattern, will handle the greater
amount of data required by the increasing size and pace of
world activities, and we will accomplish this with greater
ease. The electronic machines will largely take care of the
matters of quantity, spee.d, and in,terconnection, and the
human partners will then be able to elevate themselves to
the more difficult, truly human, intellectual tasks.
In such a world, what will happen to "management"?
\Vill lIlanagers be in part replaced by machines? Will managers have their brains "extended" by the new electronic
tools availahle? Will management perhaps need to be redefined, when the decision-making, control, and planning
fUllctions in the world of the future change in form and in
16
substance? It is such questions that we shall examine in
this essay.
We must start by describing a few more highlights on
the nature of the technological society ahead. Management,
as we know it today, will change because, for one thing, the
world will change, and not just because management itself
will be seeking to make use of new technological tools. We
must assume that everywhere we look in the environment
of the latter decades of this century we shall see new devices
for keeping track of all of the information needed to per-.
form the tasks in which men and machines are today engaged.
There exists already sophisticatiQn in such problems as
the automating of a chemical process by automatic measurement of the characteristics of the entering raw materials,
the chemical that is produced, and the temperature, pressure and other physical parameters. We may expect this
kind of application to grow in many ways, to be common
rather than exceptional, to provide an optimum operation
through a better matching of man and machine and a high
degree of interconnection and integration with other aspects of the total operation. We should expect to see a
growing corps of specialists controlling the local "on-line"
operations, with "local" and "on-line" covering more and
more territory. In addition, there will be another large
class of man-machine combinations working to understand,
design, analyze, and improve the operations in a broader,
longer range sense.
Flows and Traffic
Similarly, we may expect to observe that the flow of
people and things on the ground and in the skies, in city
and country, will be dependent on networks of devices to
sense the traffic, anticipate the conditions that will develop,
figure out what signals should go out to modify. and control
the flow so ~s to ac,hieve the greatest efficiency, capacity, and
safety, and to communicate this information to the devices
or the people that perform the operation. Again, we see
t'Yo classes of problems. One is associated with the "online" operation itself, that is, relating the developing situation to- the past and to a set of rules or equations, and, finally, originating new controlling signals. The oxher is the
analysis and design of the operation. This function will
also be a man-machine exercise whose purpose is to seek
improved operation, better utilization of the total resources,
COMPUTERS and AUTOMATION for July, 1963
"
optimum application of machines and men in a harmoni~
ous ensemble, and to better define the goals of the' operation.
Missile Test Flights
In engineering work, we already have reached the point
of great dependence upon electronics to extend the brains
of the scientist and the engineer. For example, it would
have taken us some ten thousand actual test flights to prove
out the design of an intercontinental ballistic missile without the use of electronic brains. Actually, on a crash basis,
we did it in a few years and needed only some tens of
flights. The large-scale digital computing systems and simulators enabled us to 'get the equivalent of tens of thousands
of flights. We were able to work out the important interrelationships amongst all of the design variables in the laboratory, ahead of the modest full-scale flight tests that then
confirmed the soundness of the design.
Fro'm Problem to Design
But in the future, we shall expect to go further. To do
computation and simulation more rapidly by machine is
only half of the ultimate potential. One day the engineer
at his console will be able to proceed directly from an articulation of the needs, the problem to be solved; by tapping a huge but accessible file of scientific facts, laws, and
fundamental interrelations, he will be able to arrive more
quickly at basic conceptions of design. He will be able to
compare alternatives and to know what range of solutions
he can consider. He will be able to synthesize and create
more quickly, doing a higher level of difficult thinking than,
unaided, he can hope to attain.
Again we see the intellectronics age as one in which the
activities can be considered as of two kinds. One involves
the day-to-day operations within an existing and understood mode of operation, with the man and machine combined to do a much better job of handling the matter both
as to quantity and as to rate. This covers situations that
are common and recurring. They can be planned for ahead
of time and covered by the stored information, the incoming information, and the existing logic of the system. Even
the exceptions can be identified and called out for decision
by the human operator. On the other. hand, we have the
possibility of the highly intellectual human operator using
the electronic machines as partners to enable him to analyze
the whole of what he is attempting to do, to improve his
basic conceptions and goals, to redesign the entire operation, and to explore new possibilities.
Medicine
In medicine, we see some especially strong indications
of how the intellectronics age may advance. The physician
of the future will be able to take the data on a patienthis apparent condition as judged by tests and the physician's own observations, his history, the complaints or
symptoms-and enter these into a national network of statistics and deliberative services. Almost instantaneously,
the physician might have displayed to him pertinent comments from the system concerning this patient's possible
diagnosis and possible treatments. He will get in part the
equivalent of having consulted with thousands of other
physicians. He will have pertinent facts and issues called
to his attention. He can propose a treatment and have this
criticized against the statistics for that patient's data.
A whole new profession, having to do with the statistical
handling of medical facts, relating cause and effect, drugs
and cures, will probably grow up. Considering the speed
and capacity of conceivable national intellectronic systems
in the future, such systems may have an impact on the practice of medicine comparable to the advent of surgery.
COMPUTERS and AUTOMATION for July, 1963
Again, there is the direct, immediate, day-to-day problem of the practicing physician which may change in nature
because of the existence of a suitably designed intellectronics network. Also, we see again the possibilities of new, contemplative, longer range activities in the field of medicine
made possible by the availability of a tremendous volume
of facts, of the electronic means for the analytical examination of these facts, and of !lew intellectual skills that are
brought into existence when the data and equipment become available.
Law
Similarly, in law, the attorney of the future, applying
himself to the day-to-day activities of steering the actions
of people, corporations, a,nd government along lines prescribed by the nation's laws and rules,_ can expect radical
changes and improvements. He will be able to take the
facts related to a particular action with which he is concerned-such as a business deal, a contractual relationship
between two people, or a petition for a license to perform
some activity-and introduce these facts into the legal
intellectronics network of the nation (and perhaps of the
world, in some respects). Again, almost instantaneously, he
will get back approval with confidence that what is proposed has been properly examined, is entirely consistent
with the rules, is not in conflict with the claims of others or
with the total facts as they are known to exist by the system, and is now properly recorded. Or he may receive an
indication that the case is not complete, that information
is lacking, perhaps that he has not properly portrayed the
total situation, or that he is attempting to arrange something which is inconsistent and improper, or something that
may be legal in part but that requires special called-out
a pproval according to the rules.
In more complex situations, where there are differences
to be resolved or problems that must be presented to deliberative bodies for action, the system helps to provide a
definitizing of the differences in the case or the particularities that require judicial or other action. The attorney, in
trying to see how to proceed, can present the gist of his case
to the intellectronic system and receive back immediately
a summary of similar cases. Indeed, the summary before
him would indicate whether the cases presented are truly
similar. If not, this will be notice to the attorney that he
has not properly understood his case, because he apparently
has not described the essence of it well enough to receive
back pertinent comment.
Two Sets of P'urposes
Again we see that man-machine, intellectronics systems
can serve two classes of purposes. In one, the attorney can
get help on his day-to-day, immediate operations. A manmachine system will do better in assuring that actions are
legal and proper and that all activities of the world's operations are done in accordance with an orderly, predesignated
set of rules. In addition, intellectronics can provide a fuller,
up-to-date understanding of what goes on, and this is part
of the job of improving the rules of the nation. This takes
us crom the day-to-day practice of law to the deliberative
bodies that make and judge the laws. Here, the ready
availability of all the facts and rules, and the ahility to
process th~se, allows analysis of what issu('s or situations
present problems and seem to require !lew rules, of what it
is that is creating inconsistencies and cOllllicts. The intellectronic systems of the future, with man and machine in
proper relationship to one another, will make possible the
study of needed legislation without so much guesswork and
emotion.
Through this example, we are afforded an opportunity
to observe another factor which will be a common one in
the technological world of the future. The rules of opera-
17
tion as determined by the government will gradually be
altcrcd so as to take advantage of what technology affords,
not only in the interconnected national library of facts and
speedy information retrieval, but in the processing and
examination of that information, especially in relationship
to the rules.
Covernment
New patterns for the handling of the government functions of policing, referceing, and control may be expected.
Today, most things are done without reference to any central rccording system, whether ultimately to be policed by
govcrmllcn t or not, and they are allowed to rest unless
someone suspects or complains that there is a violation.
Then activity or confusion results, and in one way or another the matter usually gets straightened out, or it dies,
or occasionally someone sees how to alter the rules to avoid
inequities or confusion. In the future, with the speed, interaction, and capacity of data collection and analysis so
vastly increased, it becomes possible to set up controls to
cover automatkally a wide variety of situations. This has,
of course, both good and bad implications. On the bad side,
we could arrive at an overcontrolled society. But if it is
done right, it means that we can have a clearer set of'rules
and a less chaotic, more orderly society. Furthermore, it
means that, in the process of deciding what rules to have
and how to get them changed, the investigations will be
performed better.
It is even possible to imagine that the intellectronic
system could be asked by the lawmakers what the situation
would have been for, let us say, the previous five years, if
the rules had been different along some proposed line.
Then, the system, having tremendous speed and access to
all of the information as to what has happened, could go
through a simulation process with very considerable accuracy, and it would turn up with comparative data showing
the consequences of changing the rules in various directions. Lawmaking and the nation's interest in lawmaking
could thus undergo enormous revamping.
Management of a Business
Let us now start to deal more specifically with problems
that have to do with the management of a business.
Day-to-Day Control
Clearly, again we shall see two kinds of problem areas.
One aspect of management is concerned with the day-to-day
control of the operation. Today, we do this by a combination of a sort of "on-line" control together with what might
be called "conventional accounting," which means that we
gather the facts as to what has happened after it has happened. In "on-line" control (and I do not refer here
mainly to thc concept of process control in a factory or
refinery), the managers, who are aware of what is happening and who havc some plans, biases, goals, or ideas as to
what ought to happen, or what they wish were happening
instead, more or less continually make decisions which alter
the operation. They arc constantly at work, sensing the
operation and changing it. Whcn it is possible to do this
all a short time basis, we can think of it as "on-line" management. Presumably, the more complete the information
as to what is happening, and the faster this information can
be made available to the sensing system of the manager, the
clearer the manager is on the relationships amongst all of
the factors, and the closer he can get to "on-line" operation.
Delay in Knowledge
If the rules of the operation are quite complex, and if it
depends on a rather vast amount of information, far beyond what one manager or even a host of human operators
can hope to fathom with their brains and senses-then the
control system is probably quite far from "on-l~ne" manage-
18
ment. To be sure, managers today are busy making decisions on matters that are brought to their attention. The
average manager would probably tell you correctly that he
is dealing with sudden, new facts every moment and having
to adjust to them, often very rapidly, and, by gosh, he is on
top 0,£ his operation. The point is, though, that a very
large fraction of all the significant things that happen in his
operation are actually made known to him quantitatively
and completely only after a substantial period of time. He
gets, in other words, an :.:~cc~unting." The main descriptions of his operation, by and large, are those that go
through the accounting process. And most of the machinery
to improve the flow of data, and to increase the capacity
and speed of processing, has to do with trying to give him
a more complete accounting and to give it to him sooner.
Rapid Knowledge
Now, what happens in the future to day-to-day management as the information quantity and speeds are drastically
increased? Suppose this in1formation flow really enables a
manager to control what
happen out of the available,
understood, up-to-the-minute spectrum of choices. We are
not talking about the local loops involved in some geographically or operationally rather isolated element of the
total operation. That is, a particular machine' in a factory,
for example, may be run by a computer and a taped program; it has its own local detailed decisions to make. It is
a form of "on line" management, and it may even involve
a human operator in cooperation with the machine's automatic observation and programming capability, with the
combination striving for some optimum utilization of the
whole setup to turn out its product. In a similar way, in
the banking business, we are not primarily concerned with
such a local problem as a combination of human operators
and machines to sort checks, insuring that they are flowing
smoothly from one end of the line to the other. We are
concerned with something that we have a right to call
operating management and more particularly with controlling an ensemble of man and machine operations so
that the system accomplishes what the managers want.
However, we do not have to go to the other extreme of
simply looking at the final result, the profit at the end of
the year, or the return on investment. There is a tremendous range of operating management problems-'above
the small local ones and yet far away from the summary
management-where better operations management is basic
to over-all improvement.
will
Over-all Control
Before we go ahead to answer the question of what
might happen to operating management in an automated
information world, let us look briefly at the other aspect
of management. Above the day-to-day operating control,
there has to be a defining of the,.goals of the over-all operation, of goals for the operating management, and an assessment of operating management including the way in which
men and machines are used. There must be an attempt to
assure a better operation by attacking the fundamentals of
the whole technique of management. As to this second category of management, we might expect it to be altered and
improved, for one thing because the operation below will
be better controlled, and the plan for the operation will be
more quantitatively specified in the beginning. It will be
a lot clearer what can or cannot be expected out of the
operation. There will be some good estimates as to what
goals are reasonable. The operation will be so managed
that it will come a lot closer to the optimum that conditions
allow than we have any right to expect of most operations
today.
Next, we should note that the operation will be tied
in more closely with other operations of the world. Integra-
COMPUTERS and AUTOMATION for July, 1963
•
PSA
PACIFIC SOUTHWEST AIRLINES "THE NATION'S LEADING INTRA-STATE SCHEDULED AIRLINE"
•
"WHY WE CHOSE THE NCR 390 COMPUTER."
PACIFIC SOUTHWEST AIRLINES, SAN DIEGO, CALIFORNIA
"The NCR 390 Computer provides us with
the fastest, most accurate system for travel
agency billing, for corporate account billing, for maintaining ticket inventory, and
for writing our payroll records.
"By improving service to our customers
and to our employees, the NCR electronic
system helps PSA maintain an edge in the
highly-competitive air travel industry.
"Information on flight tickets is automat-
ically punched in code in strips of paper
tape each day. Information about each
travel agent and corporate customer is
printed on a unique magnetic ledger card
that also carries the data in strips of magnetic tape on the back of the card. The
program of the NCR 390 enables us to post
our statements and ledgers electronically,
and keep our ticket inventory automatically.
"In addition to handling this specialized
data processing job, the NCR 390 also
handles the payroll records for our entire
organization."
J.
7~.f dJ;..--.---
J. Floyd Andrews, President
PSA (Pacific Southwest Airlines)
NCR PROVIDES TOTAL SYSTEMS - FROM ORIGINAL ENTRY TO FINAL REPORTTHROUGH ACCOUNTING MACHINES, CASH REGISTERS OR ADDING MACHINES, AND DATA PROCESSING
The National Cash Register Co,.1,133 0ffices in 1:20 countries· 79 years of helping business sllve money
COMPUTERS and AUTOMATION for July, 1963
19
tion and interaction will be much more a factor in determining the nature of operations in the future than at present. Raw materials will flow from a source. perhaps a
distant one, to the processor or assembler in accordance
with signals generated, in major part automatically, as a
result of the operating "on-line" relations between the two
plants. The shipping of a component from one plant will
ha\'e been the result of an automatic announcement of the
need for that comphnent at some appropriate future moment in another plant. In addition, a tied-in, bankingaccounting intellectronics system to which the manufacHIrer, the supplier, and, finally, the customer arc all
subscribers, will insure an appropriate and automatic charging of their accounts. Information on taxes, flow of funds,
allocation of charges in banks and to various accounts, will
all he transmitted by electrical signals, traveling through
wires or through the air, perhaps even by way of communications satellites. .-\ closely knit network of information
flow will relate each separate organization with all the other
groups with whom it has dealings, to assure that everyone
gets what he needs at the right time and the right place.
This whole chain will furnish the same kind of information to the government to insure that all is done in accordance with the rules, with an automatic calling out of any
violations.
If you are bothered by having to fill in the social security numbers on all of your little dividend checks to conform to the new rules of 1963, then imagine the period of
the future. Probably none of us, at least in urban areas, will
deal with money. Identification by thumbprint in front
of an electronic scanner will make possible the transfer of
funds, whether you buy a house or a necktie. There will
still be risks for those who live. in that beautifully ordered
period of the future, but they will be different, more
"modern," risks than the ones that we have today.
Evaluation and Control
\Ve thus are caused to observe that the higher level,
analysis, goal·searching, evaluation functions of management in the future may involve people working with consoles, calling out information from the system and studying
automatic displays of new combinations from that information, exploring possibilities, all somewhat aloof from the
momentary "on-line" operations. The full powers of intellectronics will be available to remember, to consider the
impact of modified rules and decisions, and to display these
to the operators. One output of this higher level of management analysis will be a modification in the man-machine
operating management system, both as a network and as to
the controlling parameters. The more superior this top
level job, the better will be the control arid the less the
deviations from plan in the day-to-day operations.
Let us now return to the "on-line" control management
system of the future. We see that the primary function of
conventional accounting for day-to-day operations should
essentially disappear. The system becomes one which generates detailed plans, observes the deviations from plan as
they occur, and takes action regarding these deviations. To
some deviations, it reacts in accordance with plans for deviations from plan. For instance, suppose that a factory's plan
is upset because a machine has had some sort of mechanical
failure, or because workers fail to show up due to a sudden
epidemic. Either the human managers are at this point
brought in to make a decision amongst a group of alternatives, or they have already made a decision as to what will
he done when these things happen. In the latter instance,
it will be only when there is an even more "unplanned"
and extraordinary deviation from plan that will require a
human managerial input.
20
Optimizing Over-all Performance
\Ve should mention briefly another rather important aspect of man-machine relationships in the future. Always in
creating a system for control, we will have available as possible components the human mind and sensing system and
electron~c machines, either in existence or capable of being
developed. A good system is one that so combines and
mixes these as to optimize the reliability and over-all performance against the investment. There are many functions which the human mind and sensing system can do
better than machines, because the human mind has the
equivalent of some 'billion or more transistors in peculiar
interconnections and has its own built-in power supply and
is readily transportable. There are other things for which
the human mind and sensing system are decidedly unsuitable. Thus often we would expect the best systems configuration to co~bine man and machine. But for the purpose of this discussion, we don't have to know the division
of labor between man and machine, or the exact configuration of the system. We can assume that the man-machine
system of electronic devices and human operators has the
job of "on-line" control. ,This includes generating an operating plan, observing deviations from plan, introducing
actions when such deviations are observed, and automatically generating new plans that are better-all in response
to programmed goals and over-all assessments made by the
top level of management.
Technology will make possible, as time goes on, the
economical handling of a hugely increased amount of information that relates to any given operation. Furthermore,
technology will provide economic means for processing and
bringing the information virtually instantaneously to a
human operator, wherever he may be, and to machines,
in such form that the men and machines will be able to
understand and use the information when it arrives. All
this is basic to understanding the operation better so that
it can be optimized,~s well as being basic to on-line control.
But simultaneously, 'we can abandon a good deal of other
information that is now thought to be needed and is really
concerned with conventional accounting.
In the future, as we develop systems that are able to
generate optimum plans, we shall be interested much less
in the normal daily details and much more in the deviations, and especially in the deviation-reacting system itself.
We become less interested in what has routinely happened
as we become more certain that what happens is the best
that can be caused to occur. We will be more interested in
understanding what we mean by "the best," and in improving the system for determining and adhering to this optimum. Operations will become more optimum, more quantitative, more understood in the sense sometimes called
"scientific management" in the past. Management will be
occupied increasingly with the goal of a truly "controlled"
operation, with pushing forward the frontier of optimization, and with quantitizing the parameters of the operation
to make these things possible.
The operations of the world can be expected to become
more integrated, faster paced, with greater interactions. In
that complexly interconnected world in which government,
industry, the buying public, the money-accounting-tax-legalpolicing elements are all tied together in a complex, tight
network of electronic messages, cables, and computers-in
that highly technological world of the future-it does not
look like a very good place for the maverick or nonconformist. The isolated, separately and personally run operation would appear to find itself with difficulties in tying-in
as it would have to according to the rules of the nation of
that time. It would be almost like trying to bring a horse
and buggy onto a fast-moving, high traffic density freeway.
COMPUTERS and AUTOMATION for July, 1963
.J
,
You might be able to get to the on-ramp but you would
ne\'er be able to get in between the cars in the first lane.
which in principle arc actuafly beneath the human brain,
that the human intellect will be freed to rise to the higher
tasks of life, lea\'ing the lesser tasks to. his machine partner.
Originators and Creators
What then will the originators, the creators, the imaginative entrepreneurs in the managerial field do in that
society of the future? Well, surely, if anyone really sees a
problem here, then part of what I hoped would be apparent
between the lines has been obscured by the speaker. Because, certainly, as information handling capacity, speeds,
and interconnections grow, and as operations become more
controlled and optimized, the increased number of parameters that determine the state of the business will make much
more difficult the soul-sear-ching for a better description of
the goals. What will be asked of a manager, now that he
has all of this information avai~able, will be a more challenging task even than the task technology faces in providing the new tools. It will be like providing the biologist
with such magically, vastly increased magnification for observing living matter that he becomes aware of many more
forms of harmful viruses and more new diseases for which
there are no cures. In time these new problems will be
solved, presumably, but on the way the opportunities and
possibilities will increase more rapidly than we will have
the brainpower to exploit.
It must always be the case, with the unlimited mysteries
of nature, that the new tools and skills given to the human
mind for exploration will increase the area of exploration.
The challenges for the imaginative genius will be there
in greater measure. Perhaps what information-automation
will largely do is make possible such efficient handling of
the mundane, high-quantity, high-rate intellectual tasks,
(Based on a talk before the 5th SYlllposium
ou Process A utoll/alioll. A jJril, 1963)
~C~df
omputer, computer, on t he wa11,wh'
0 S t he f'
auest one of all?"
Programmers are cordiallyinvited
to discuss interesting career opportunities
at all experienced levels
in the IBM Data Systems Division
Immediate assignments are open with
programming research, development, and applications teams in:
•
Advanced programming
Programming languages
Advanced programming techniques
Business-oriented programming
Supervisory programs
Automatic Operator
Machine Control
Systems Supervisor
Stack Job Scheduling
Symbolic I/O
10CS
Interrupt Control
Symbolic Debugging
IBM benefits include:
Advanced-degree program
Relocation expenses
Locations are in
Poughkeepsie and New York City, N.Y.;
Beverly Hills, Calif.; Boston, Mass.
Promotion from within
If you have one or more years of experience, please write, outlining your back.
ground and qualifications, to:
J. B. Zwynenburg
IBM Corp., Dept. 539G
Box 390
Poughkeepsie, N. Y.
IBM~
IBM is an Equal Opportunity Employer
COMPUTERS and AUTOMATION for July, 1963
21
READERS' AND EDITOR'S FORUM
(Col/til/lied from Page 7)
cipally from the United States and Great Britain. ACM
conferees will also tour the United States Air Force Academy at Colorado Springs and the National Bureau of Standanls scientific laboratories at Boulder, Colorado.
A panel discussion of significant topical interest will be
held on "Social Aspects of Computing: The Responsible
Use of Computers in Politics." A number of provocative
questions will be raised: "Does information provided in
computer political analysis affec,t the integrity of the candidate in his campaign? Can highly sophisticated computer
analysis so 'manipulate' the electorate as to constitute a
violation of iildividual rights? Can computers have a programmed morality by feeding into them information on
history, philosophy and religion?" Dr. Edward Bailey, professor nf, psychology at the University of Colorado, will be
chairman of the panel. Other discussion leaders will be
Dr. W.illiam McP~lee, University of Colorado, Rev. Harry E.
Hoewischer, RegiS College, Denver, and Dr. Louis Sutro,
Massachusetts Institute of Technology.
A number of new techniques are being introduced in
the application of computers in medical diagnosis. Intricate
problems are involved, not only in the collection of medical
data but also in "mathematical model building"-the basis
for computer "decision making" when the machine is faced
with a diagnostic problem. Dr. Robert S. Ledley of the
National Biomedical Research Corporation at Silver Spring,
Maryl.and,. will. lead a panel discussion on this subject.
P~ne.hsts ~nclude: Dr. Theodor D. Sterling, University of
ClllClllnatI; Dr. Clifton F. 1\Ioutain, University of Texas;
Dr. Caesar Caseses, U. S. Department of Health, Education
and Welfare; Dr. Stanley Woodson, Lovelace Medical
Clinic, Albuquerque, N. 1\1.; and Dr. Joseph Balintfy, Tulane University.
Papers in 14 subject categories will be given: logic, compilers, numerical analysis, pattern recognition, mathematical pro.~ramming, bio-medical programming and processing,
educatIOn and programmer training, information retrieval,
hardware, programming languages, software, language and
learning, simulation and graphical output, and merging and
~or~ing. Th.e 33 sessions include 80 contributed papers, 8
InVIted papers, 7 panels, 3 halls of discussion and 3 evening
tutorials. Those attending will be able to secure abstracts
of all papers in advance of the conference.
Mr. William C. Norris, President of the Control Data
Corporation of Minneapolis, will give the keynote address,
"The Computer Industry-A Look Toward the Future."
Dr. Alan J. Perl is, director of the Computation Center
and professor of mathematics at Carnegie Institute of Technology, Pittsburgh, Pa., is current ACM president. General
Chairman .of t~e conference is Dr. William H. Eichelberger
of the UnIVersIty of Denver. Fred P. Venditti, chairman of
the technical program, is also of the University of Denver
host institution.
'
ACM MEETING FEATURES
"WILD WEST" TOUR
An example of Western hospitality, Colorado-style, is in
store for people attending the 1963 ACM National Confer~n~e ;~nd International Da.ta Processing Exhibit August
_7-JO III Denver. W. H. EIchelberger of the Denver Research I nstitute, conference general chairman, reports a
"colorful" program is in the making.
A highlight of the conference's special events is a big
round-up to be staged at East Tincup, Colorado's recreated
old mining town, done in the style of the West in the 1880's.
Among the sights for computerites are: gas-lighted streets:
watering troughs, blacksmith's shop, sheriff's office and jail,
22
Boot Hill, the hangin' tree, "Old Tombstone," and gunfighters in action. The evening will also feature a Western
Chuck Wagon Bar-B-Que and entertainment with Western
songs a nd stories.
Several field trips are scheduled during the conference.
One is to the United States Air Force Academy at Colorado
Springs. Another is to the National Bureau of Standards
at Boulder, including a look at the Colorado University
campus. Tours of Martin-Denver will also be arranged
during the conference. Activities for wives will include
coffee hour socials on each day of the conference. Tours
will be available of Denver, the mountain area, and Centtal City.
OBSOLESCENCE OF EXPERIENCED
ENGINEERS AND SCIENTISTS
Careers Incorporated
New York 21, N. Y.
A study recently completed by Careers Incorporated and
published in "Personnel Administration" indicates the existence of an alarming number of "obsolescent" engineers
and scientists in the United States.
The study, which surveyed registrants at Career Centers
in four metropolitan areas, revealed that 54 per cent of
degree-holding registrants failed to receive a single interview bid from any of the defense contractors represented at
the Centers. An average of over twenty major defense contractors had representatives at each of the four Centers
surveyed.
The published text of the survey states: "There would
seem to be considerable substance to the theory that our
so-called technical manpower shortage is really just as much
a question of proper utilization of our engineers and scientists ~s it is a question of a purely numerical shortage."
In an analysis 0f the engineering fields exhibiting the
greatest degree of "obsolescence," the study revealed that
civil engineers were the least in demand, with 91 per cent
receiving no interview bids at all; 64 per cent of the chemical engineers and 51 per cent of the mechanical engineers
similarly received no bids for interviews.
By contrast, 27 per cent of the data-processing specialists
received bids from five or more employers as did 18 per cent
of the physicists and 23 per cent of those in electrical and
electronic sciences.
Another finding of the study was that 17 per cent of the
degree-holding . registr~nts were unemployed at the time
they came to the Career Center. This figure is surprising,
inasmuch as it has bee,n generally assumed that virtually all
engineers and scientists are employed, and that those who
are job hunting are interested in changing jobs, not in
simply finding jobs.
The Oareers study underlines tJhe recent emphasis that
has been placed on obsolescence of scientific and engineering m for full rerun() if not satisfactory (if
in good condition).
My name and address are attached.
;;;;;;;;;~f1111111111111111111111111111
I I I I 1.1 ... 1 '1"""
I •••
1.11 I I I I I. I
I.'
II I I I 111.11 ...
23
-
CONTROL DATA® 3600 Computers Selected for
Real· Time Computer System
The Atlantic Missile Range will place in operation a dual CONTROL DATA 3600 real-time computer system
to provide data for range safety at Cape Canaveral. High internal speed in the 3600 compute module is
matched by the speed with which the CONTROL DATA 3600 can accept information from multiple sources,
decide which is most accurate, then compute and transmit the results in the form of a missile impact point
every 50 thousandths of a second. 0 The major elements of the Real-Time Computer System are two
standard CONTROL DATA 3600 computers, the world's most powerful computers commercially available.
Along with its high speed, another major advantage of the CONTROL DATA 3600 is the provision for modularity built into the machine. Additional memory modules, compute modules, and input / output data
SALES OFFICES: Albuquerque • Beverly Hills. Birmingham • Boston • Chicago. Cleveland. Dallas. Dayton. Denver. Detroit • Honolulu • Houston • Huntsville • Ithaca • Kansas City
(
v
•
Atlantic Missile Range
channels can easily be included to handle increased data processing requirements as needed. This means
that the Atlantic Missile Range will start out with a system commensurate with their requirements. As
AMR's data processing needs grow in size and complexity, additional units can be included without replacing the original equipment. D Control Data's total system capability for this real-time system is provided
by the Company's Government Systems Division .. .and includes problem analysis,
system design, hardware implementation, system integration, programming, instalCONTROL DATA
lation and maintenance. D To learn more about the 3600 and Control Data's total
CORPORATION
system design capability, contact the Control Data representative nearest you. 8100 34th Ave So" MIIIIII~.lpohs 20, Minn,
• Los Altos. Minneapolis • Newark • New York City, Norfolk • Orlando. Philadelphia, San Diego, Seattle. Washington, D.C. / Luzern • Zurich. Uad lIorntJurll •
1'..,,, • M"liJour"u
PROJECT
TIME/COST
ESTIMATED
MACHINE TIME
ESTIMATED
FUNCTIONAL
PACKAGE
APPROVED
PROJECT
TIME/COST
ESTIMATES
REVIEWED BY
CUSTOMER
TIAL SYSTEM
SIGN APOVED BY
STOMER
CO
PR
DE
DOCUMENTA TION
REQUIREMENTS
DETERMINED &
APPROVED
CONTRACTUAL
REQUIREMENTS
DETERMINED
FUNCTIONAL
PROCESSES
DEFINED
Russell D. Archibald,
Consultant
Hughes Dynamics, Inc ..
Los Angeles, Calif.
Management information and control systems offer a leading application of computers in the next decade. A major element of
this application is the use of network planning techniques such as PERT. The author
provides an instructive analysis of the background, development and use of computerassisted PERT programs.
PERT (Program Evaluation and Review Technique)
has survived the critical scrutit:ly of realistic managers and
is here to stay. Managers have found that PERT:
• Provides a way for planning on a uniform and logical basis.
• Provides follow-up assurance that planning has been
done.
• Permits plans to be kept current.
• Lets management foresee the impact of variations
from plan and take corrective action.
Background
The need for improved planning. and progress evaluation in modern "massive engineering" projects became apparent in several industries-construction, process and defense-at about the same time. Evolution of the network
plan or arrow diagram concept, which is the backbone of
both the Critical Path Method and PERT, was a fundamental step forward in the development of better management information and control systems. It is illustrated in
Fig. I.
The CPl\f arrow diagram network evolved from detailed
Gantt bar charts which were job oriented. Linking jobs
together in dependent sequence produces the arrow diagram. often without identification of connecting points.
The PERT network evolved from a combination of bar
charts with milestone charts (milestones are defined as special events, or instantaneous occurrences in time, of interest
to manag(,ment). Milestones are useful for progress evaluation, to determine if a job, represented on a bar chart by a
long line, is ahead, behind or on schedule.
26
PERT AND THE ROLE OF
Network Elements
Two elements make up the network or arrow diagram:
(I) the line or arrow, representing time-consuming activities
or sequential constraint; and (2) the circle or rectangle, representing the beginning or end of an activity, or a milestone. The most widely used and accepted names for these
are: (I) activity and (2) event.
An event is defined as a specified accomplishment (physical or intellectual) in the program plan, recognizable as
occupying an instant in time. Events (also called nodes)
consume neither time nor resources and arc represented in
the network by circles or rectangles.
An activity is defined as a time-consuming element in
execution of a task. It is represented on a network or flow
chart by an arrow. Events are separated from one another
BAR (GANTT) CHART
--=---
=I
CPM'/
CONSTRUCTION AND
PROC!':SS INDUSTRIES
I
I
J
I
PERT
~IILITAHY & SPACE
R& D
lIIOIlE DETAILED
BAR CHARTSJOB ORIENTED
I
I
COMBINED BAR
AND MILESTONE CHART
___
I
EVENT
ORIENTED
_=--.1
PERT NETWORK
CPIII
A~M
\
NEXT GENERATION
NETWORK BASED
PROJEC T-IIIANAGEMENT
SYSTEM
WITH BOTH
EVENTS AND ACTIVITIES
IDENTIFIED
Figure 1 - Historical Evolution of the Network Plan
COMPUTERS and AUTOMATION for July, 1963
NAL
:0
~ACHINE TEST
TIME AVAILABLE
SYMBOLIC DECK
CORRECTED
RERUN
TESTED
PROGRAM
REASSEMBLED &
FINAL LISTINGS
MADE
MACHINE DECK
CORRECTED
MACHINE TEST
COMPLETED
'LE OF THE COMPUTER
by actIvIties, and an actIvIty cannot be started until its
preceding event has been accomplished. Likewise, a succeeding event to an activity cannot be accomplished until
that activity (also called task, job, arrow) is complete.
The final step in a network plan, as shown in Fig. I, is
addition of interdependencies between elements of effort
which may be called constraints or dummy activities. Constraining activities may consume (I) insignificant time or
resources or (2) significant tillle and essentially zero resources. Several practical difficulties are avoided if such
constraining activities are so defilled that no resourccs are
expended on them, since they usually reprcsellt a trallsfer
of paper, information or hardware from olle orgallization
to another.
Further refinement of the plan is achieved by adding
more detail to the network. Long activities are broken up
into short elements of efIort, so that meaningful activity
time estimates can be made. This requires definition of
more events to correlate the beginning and end of the activities. Concurrent with this addition of detail, more interdependencies or interfaces may be identified and placed on
the network.
A network plan is defined as a graphic portrayal showing time dependencies and the chronological sequence of
events and activities leading to given end objectives.
A typical simplified network plan, showing a portion of
an over-all computer installation program, is illustrated in
Fig. 2. The broken lines indicate constraining "zero time"
activities, and the heavy solid lines show the longest or
"critical" path.
Use of the Network
The network plan is a basic management tool which can
he used in a number of ways.
A. Time analysis of plans is accomplished through the
following steps:
(I) Estimate activity time spans
(2) Add activity time spans forward through the
network to determine "expected" dates
(3) Subtract time spans from the end schedule date
to determine "latest allowable" dates
(4) Subtract "latest allowable" dates from "expected"
dates to determine "slack" time (allowable slippage)
(5) Locate "critical path," the longest series of
COMPUTERS and AUTOMATION for July, 1963
events and activities through the network (this
is the "least slack" path)
B. Integration of plans by linking together two or more
networks.
C. Progress Evaluation through reporting of actual completion dates which are used for the next time analysis cycle.
D. COlli lmta j))"()c(?.I'.I'ing of the network is common be. cause of the Iletwork's unique adaptability.
E. l.fll)()/" lind ('(Jsl (',llil/llllin,!!,' alld correlation to actual
Iahor alld ((1st call be aC((llllplished with the PERT /
COST Iletwork.
F. Sill/Illation find opti1l1ization of project plans can be
performed using the network as the project model.
Role of the Computer
The computer has been linked almost synonymously
with PERT since its initial application to the Polaris missile program. The Critical Path Method, however, has not
placed such reliance on the computer, instead encouraging
manual network analysis methods. This is due largely to
the difference in network size, the resulting mass of detailed
PERT data in the typical defense industry application rcquiring a computer to llandle the volume.
Thc network plan, with its events and activities, is
uniquely adaptable to electronic data processing. By merely
assigning numbers to events, the entire network structure
can be fed into the computer alld all arithmetic, statistical
probability calculations, alld logical operations can be performed. This adaptahility is actually one of the major features of PERT for hetter planning and progress evaluation.
Not always do network plalls require a computer, how··
ever, to generate desired analysis results. In many cases, it
is cheaper alld faster to calculate the various parameters by
halld. Beyolld a given size network or set of networks, however, this hecomes impossible. Just where the breakpoint
occurs callnot be set by a simple rule, but depends on number of clements in the network, up-dating frequency, numher of changes being made to the network, availability of
an operating PERT computer program and other similar
factors. Detailed study is needed before recommendation
can be made for a particular application.
The computer brings a variety of benefits and capabilities to PERT:
(1) Speed: calcula tions performed thousa nds of times
faster than by a human
27
.f1
q
oi'l-_se_c_ur_e_D_a_ta_ _ _*t'15>--_:::-T-;ran-,-sc-:r:--;ih=-e..,..F_ir_s_t--f'\16 Transcribe Remaining Data 29
Collection Forms
Batch of Data
from Records and Files;g
'1'(5)
(22)
~
'"
ff !:
~
(5)
J
f~:
"f
o§,cJ J§;j'1
~ r
::
Je
£
9
"l
g
Parallel Operation
31}-an--:d;-:Sy~ste:-m---'A:-ud-:"1,","Ot--<32
(22)
System
Operatiorial
(D) = Duration in Working. Days - Estimated
- - - - -........ Critical Path
Operating Procedures 22
(22)
Figure 2 - Typical Critical Path Diagram, Showing Part of Computer Installation
(2) Accuracy: errors in calculations are essentially zero,
much less than when performed by pencil and
paper
(3) Large volume capability: very large networks (typically up to 5,000 activities) can be analyzed on current computer programs
(4) Legible results: computer prints legible results at
rates of more than 600 lines per minute, eliminating
burden on typists
(5) Partial interpretation: computer can select, edit, rearrange, summarize, compare and apply rules, providing partially interpreted results with large savings in manual efrort; graphic charts and reports
can automatically he prepared, eliminating manual
art; an example is the common practice of printing
results of network analysis in four or more ways:
a) by slack path
b) by organization code
c) by event number
d) chronologically (by expected, latest, or schedule dates)
Computer Pro'blems
Benefits described are not always easily obtained. A
number of problems may be encountered, including:
(I) Unfamiliarity with electronic data processing; unfortunately, an aura of mystery has developed (perha ps purposely) around computers and their use
(2) Computers seem expensive on a per-hour basis; care
must be taken to compare true equivalent costs of
network analysis by hand and by computer
(3) Proper computer is not always available
(1) Cost of preparing a new program for a given computer to analyze networks is usually very high
(5) Some speed gains may be offset by long administra-
28
tive delays in getting data into and out of the como'
puter
(6) A poorly designed computer program can be the
source of great frustration °and inefficient procedures
P'rograms
Computer programs for network analysis differ greatly.
rhe size of the network is an important characteristic. The
maximum network size which can be handled will vary
from around 200 activities to 12,000 activities (largest operational program now available). This is a function of
computer size as well as of the program itself.
The event numbering method used in early programs
and in current ones for small machines, requires that events
be numbered in ascending order (although not necessarily
in strict sequence). Many later programs do not have this
requirement, and events can be assigned random numbers
without regard to network sequence. Random numbering
is an advantage in large networks, since it is often difficult
to maintain sequential numbering if large changes are made
in a network. Resulting re-numbering of an entire network
is a costly process.
Input formats and coding schemes are generally quite
consistent, as well-designed input forms can reduce errors
and save time and money. Output formats, on the other
hand, in layout, legibility and usefulness will vary widely
with different programs.
Wide differences exist in the processing efficiencies of
various programs, even for the same computer. This can
be significant in machine-running time and cost over a long
period.
••
Co•
O{g
J
c
tl
A
il
b
Input D'ata
s
In a typical PERT operation, activity time estimates are
written on the network plan, usually on the line representing the activity. Numbers are written on the circle repre-
I;
COMPUTERS and AUTOMATION for July, 196,
s
-
senting each event, and descriptions of events and/or activities are prepared. This data is then written on input forms,
designed so that punched cards can be prepared directly
from these sheets. Program control cards are prepared as
necessary to precede input data. These tell the ma~hine
certain variable data and indicate that input data follow.
questor's identification, type of run (PERT in this case),
and an account number for charging computer time and
cost. This request is thcn logged in, and the job is tun in
turn with other work. On completion of the run, the input
card deck is returned together with copies of output reports.
Transaction Codes
Wide variation of output reports is to be expected, the
result of the desire to print out most meaningful results for
a given application. Basic results of a time analysis of the
network plan are contained in essentially all the various
types of output reports. Differences relate to the optional
items of information, arrangement and legibility of the information, and types of sorts or methods of arranging
results.
Output results can be sorted by any of the items of information given. A common type of output list is sorted by
slack path. All activities having equal slack are grouped together, usually arranged by event number within each
group or slack path. Typical output of this type, from the
Lockheed IBM 7090 program, is shown in Fig. 3. Sorting
may also be done by event numbers; by expected completion data, or by organization code number. Programs have
been prepared recently which generate lists related to work
breakdown structure of the project. Graphic bar charts,
milestone charts and cost curves are also prepared by some
programs.
Most current computer programs provide results relating to each activity in the network, and a few also provide
event reports. Both types are important and useful for different purposes. The early (1959) Aerojet-General IBM
704 PERT program, probably the first operational PERT
computer program, provided both types of reports. Current
IBM 7090 programs prepared hy hoth the Naval Weapons
Common practice, following the initial Navy input format design, uses column I on the activity data card as a
transaction code. Numbers used at present are not completely standard, but basic code types include: new data, revised data, completion data and deleted data. Other special
codes are also used on some programs. Specific details of a
particular program should be studied when its use is contemplated.
The common method of analyzing a network plan is to
punch a deck of cards, one or more for each activity in the
network, and in some cases, one for each event. This deck
thcn represents the network to the computer, and is fed to
the machine when an analysis is desired. When activity
completions are reported or changes made to the network,
it is necessary to punch new cards, search through the entire
deck, pull out the old cards, and insert revised ones. This
can be laborious and time-consuming. A better method,
widely used, is to record the card deck on magnetic tape.
When changes are made, a few new cards are fed to the machine, which searches for the proper record on tape, makes
the change indicated, and the network is then ready for a
new analysis.
Assuming that a computer is available with an operating
PERT/CPM program, and that an input deck has been prepared, it remains only to obtain the actual computer run.
This is typically accomplished by submitting the input deck
with any required control cards to the computer operations
office with a run request. This will usually call for the re-
Olutput Reports
(Plt:lI.w turn to Page !lO)
PERT SYSTEM
PAGE
RUN
7
ENDING EVENT 0000000-134
BY PATHS OF CRITICALITY
CHART HP
HUGHES-PERT DEVELOPMENT PLAN
EVENT
PREDECESSOR SUCCESSOR
r
n
s
i,
It
o
NOMENCLATURE
DATE 07-10-61
DATE
DEP.
EXPECTED
ALLOWED
DATE
SCHD/ACT.PROB SLACK
EXP.
TIME
VAR.
EXP.
2.0
2.1
4.1
5.0
6.1
9.3
11.3
13.0
14.4
17.6
18.0
·20.8
22.0
22.8
24.8
27.0
32.0
38.0
.1
.1
.2
.2
.3
.5
.6
.3
.9
1.1
.4
1.4
.6
1.5
1.6
.7
.8
.9
0000-001
0000-010
0000-010
0000-011
0000-012
0000-013
0000-016
0000-018
0000-014
0000-113
0000-117
0000-110
0000-121
0000-119
0000-125
0000-130
0000-123
0000-127
0000-131
0000-010
0000-012
0000-011
0000-013
0000-014
0000-016
0000-018
0000-113
0000-110
0000-117
0000-121
0000-119
0000-125
0000-123
0000-130
0000-134
0000-127
0000-131
0000-134
PERT FICTIOUS ACTIVITY
DEV BASIC PERT COST EST FLOW DIAGRAM
DEVELOP NUMBER SCHEMES AND CODES
AGREE NO SCHEMES-CODES FOR BASIC PERT
PROGRAMMER ANALYSIS OF FLOW DIAGRAMS
STUDY REQUIREMENTS OF EXCEPTION REPORT
DEFINE AND DESCRIBE EXCEPTION MODEL
AGREEMENT ON EXCEPTION REPORT MODEL
PROGRAM BASIC PERT WITH COST EST
FORMULATE ALGORITHMS FOR EXCEPTION RPT
FORMULATE ALGORS FOR ORGAN RPT
ADD COST ACCUM SYS TO BASIC PERT PROG
FORMULATE ALGORS FOR MANPOWER SCHED RPT
PROGRAM SUMM-CONT-EVENT ORIENTED OUTPUT
STUDY FLOW DIAGRAM RQMTS MAN-SCHED RPTS
ANALYZE FLOW DIAG MANPOWER SCHED RPT
PROGRAM EXCEPTION RPTS INTO HUGHES-PERT
PROGRAM ORGAN RPTS INTO HUGHES-PERT
PROG MANPOWER SCHED INTO HUGHES-PERT
07-24-61
07-25-61
08-08-61
08-14-61
08-22- 61
09-13-61
09-27-61
10-09-61
10-19-61
11-10-61
11-13-61
12-02-61
12-11-61
12-16-61
12-30-61
01-15-62
02-19-62
04-02-62
12-16-61
12-30-61
01-15-62
02-19-62
04-02-62
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
0000-121
0000-126
0000-126
0000-130
STUDY FLOW DIAG RQMTS ORGAN RPTS
PREP FLOW DIAGRAM MANPOWER SCHED RPTS
11-10-61
12-08-61
11-18-61
12-16-61
+ 1.2+ 17.6
+ 1.2+ 21.6
2.1
2.3
0000-117
0000-122
0000-122
0000-126
STUDY FLOW DIAG RQMTS EXCEPTION RPT
PREP FLOW DIAG ORGAN RPTS
10-19-61
11-09-61
10-28-61
11-18-61
+ 1.3 + 14.4
+ 1.3 + 17.4
1.9
2.0
07-10-61
07-24-61
07-25-61
08-08-61
08-14-61
08-22-61
09-13-61
09-27-61
10-09-61
10-19-61
11-10-61
11-13-61
12-02-61
12-11~61
~
A07-10-61
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
.0
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
'0
1-
Figure 3 -- Typical Slack Path Sequence, From Lockheed 7090 Program
a
,ty
COMPUTERS and AUTOMATION for July, 1963
29
PERT AND THE ROLE OF
THE COMPUTER
[Continued from page 29]
Laboratory and the USAF AeronautIcal Systems Division
also provide both types of reports. Numerous programs for
large, medium and small computers are now available. The
latest of these provides a wide variety of tabular and graphic
(har-charts, milestone charts, networks) outputs.
Control by the Manager
\Vhether or not a computer is used, application of
PERT/CP~J may be weak and ineflective. The causes of
this are centered on the control by PERT specialists. To
get PERT away from the specialists and into the plant,
management must take control of its application.
This can be accomplished if the manager:
• Understands the basic principles
• Supports the objectives
• Ensures the validity of the input data
• Uses the results for decision making
• Recognizes the limits of the technique in scope and
manner of application
\Vith an understanding of the basic principles, it is up
to the individual manager to do the rest.
Command Systems profoundly influence the outcome of
military missions. Successful operations require systems
which provide military leadership with information to
make decisions . . . with communications to transmit
commands.
The pace of modern warfare ... vastly different from
Gettysburg ... requires computer controlled systems such
as the Air Force Command and Control System 473L.
TECH/OPS programmers are playing a key role in
providing the complex programs to make this System
operational. When completed, 473L will supply Air Force
planners and decision makers with vital information
needed to make timely and accurate decisions for any
military emergency.
TECH/OPS work on 473L is typical of the Company's
work in the System Sciences . . . CORG, OMEGA,
COMSAT, TRAG, VALOR - to name a few other
programs. Programs which have a direct influence on
military and government planners and decision makers.
Positions are available at TECH/OPS in the Washington,
D. C. area for experienced Operations Analysts and
Computer Programmers. Write Mr. J. Pierce Jenkins.
TECHNICAL
OPERATIONS (jj:Ch!OPV
Research 3600 MStreet, N. w., ~aShington 7, D.C.
I
PERSONNEL SPECIALISTS
Missiles and Aerospace
Electronics Systems and Products
Data Processing
Nuclear
Industrial
EXECUTIVE SEARCH
Salaries commensurate with senior status and
degree of contribution. Send resume or write
for our Professional Information Form. Client
companies assume all expenses.
VALLEY CONSULTANTS, Inc.
716 YORK ROAD • SALTIMORE 4, MD.
AREA CODE 301 VALLEY 5-0256
Designed to SerV6 the Professional
•
• • in a Professional Manner
An Equal Opportunity Employer
30
COMPUTERS and AUTOMATION for July, 1963
t
~rn~ reports on the field of applications programming.
Who trains computers for new jobs?
1,
The program that a computer follows in doing its work is
a logical series of simplified directions. To develop these,
the programmer must thoroughly understand the problem
he wishes the computer to solve. IBM has studied its customers' problems diligently and has worked out families
of applications to which general program systems may be
most efficiently applied.
In an unusual example of applications programming,
IBM assisted the U. S. Weather Bureau in programming
a system for global weather simulation on an IBM
STRETCH (7030). The computer program is based upon
a mathematical model formulated by the General Circulation Research Laboratory at the Weather Bureau, for
research on the problems of long-range forecasting. In
this massive system the basic processes of weather are
simulated for the entire globe in a more detailed and
fundamental manner than ever before. The simulated
weather is calculated for as many as 10,000 grid points at
each of nine atmospheric levels and for time intervals as
small as five minutes, so that over ten billion calculations
may be required to simulate the 'weather for a single day.
Even in the highly efficient STRETCH language, over
15,000 instructions were required for this versatile system,
which incorporates such varied factors as radiation, turbulence, clouds, oceans, mountain ranges, and forests.
The breadth of applications being studied by IBM is
demonstrated by these current projects: aerospace, airlines, banking, biomedicine, brokerages, public utilities,
railroads, steel industries, and warehousing. If you wish to
look into the opportunities open at IBM, an Equal Opportunity Employer, write to: Manager of Employment, IBM
Corp., Oept.539G, 590 Madison Ave., New York 22, N. Y.
··:::::::;;;Iill:::····,,,.
,." ,,·,:1:: ;;. ,,, .. ,.""."., .,,' """.,,,,,", ". "".,,,,,. "" .,"
.::::::.1
.I:::::.
"CO" ."",
~; ;
:::::: :::::; :::. ",::;::::
::: :: :~:'. :::::::::
"lIoo.ooro.oooo.OOOO.O'H.. O..... OUu.OOOO.OUUIl.~~~~:Z!~~:~~!:'''\I)')'UU(lu.uuuU'O()UU'CloO''''~OjO,uooo,,,,ouu,OIUI."r,I.Q.I,j'OU'!/lIJ"".'UJ"",. """'0," 'In'"""""
::::::::::::::
,JO.Oooo .00llU'ltOOHIOC,J·or 00' fJ'IOU,oouo'lluoo'(JU(!~~::g!gg "" ",1,,11,1,,0.1,11,11,1(,1' 1,1,,(0(,1' UUOU'I,IUUU .u"'''' .... ~~~J. U ,(,10' ",,, "OJ " '·U·U ''', ,u ... • ~~ ". " II" ""',I
"
I
II ,II, ,,,,,, '" , , ,,'
.:)00000000 'OOOO'ooooonO(lo'oUIJU'UIlOOOOOOO'OUOQ'OIlOO'OOOQ'IlI,IOOOl,,,O"'l,Iuuo.o",uu, UO HI' UUU~'OOUO'OUIIU'UO\lO' U~~~:~~~~. 3 ,0 ': :'i.'~ ~:'~~ ~I' , , ' ,. I" 'J • 'j' v',, " .. ,,,,"
:i:1 i:: :::' :::'::
.:::::::::::
,.
II
lm~jf ,:: ': :, '::1!111llll!1 ,:'. :'::!!:~ ~;: ii::.
.::::::;::::
0,
...,.:,. :"::;::
;::
i:::'.t;t ;:::.:. U . \
1/ '~.'.::' •• ;
I'
• • ""~.~ ..
,... ~
I.
1\.: oil • I', ., ,. , '0" ".',
~
,', ,', 'I' "" " " : ~d;:;;~'
""'" ", I "',,'',
::::.1
:::::::
~.~:~~:~:~.:'
, •••
~.:
~:~
'.
"~I ,.
I
"
! ~:: ~
.:~: ::~;:'
~'
" ."
;.:~."
";~' ~~
~,
I'
" " ' . . , ..
,.',.' , ,•. , I '.'.
I
':~:,
.o~::::::::::: ::::::::::::::::::::::::::::::::::~;mlm!l~!iml~ll;;;~::::::: ::::::::::::::::::::1w!:;j1lil;i~~!iiii;i;~i1:: i!lij: :::: ::::::, ::.:.:: .:::::. . :: :,:::::. ::: ::,. :::.":.:':::: ~:, :,1! .~i: :\1i ~i~:!~i ;i;~;;!~:~!;:;: ~: ;:. ::~: ': ; illll!ll1~ L
JooOO)OoOOOooOOOi)
.0000.nooo'0000'OOQO'OOOO'0000'000~:::: 0'01 ::::::::0 '''~H::r 01.00' uOOO'UJOO' )OU::UU::H:H::H::~ L:~' ~Z): ~:;~r:
,OOO'J"'" JOOOIOOOO 000001000000000'0000'0000' 0000 .0000:
rr ;:. , /),""
~ ~r
~ :~(,o'00UU'lII'C'I' U('fl" IIu"I" <)01)0. 000,,·, "f"" III 0'" 00')""'1'","
~~~~. 0 I :H~~;~~ ~~ I" :~Hh ~~OO 0 OO~U :~~~~ :~~ ~~ :~~~i !~~~!: ~ ~~~: ~~~~: O~~J' ~1?'~ ~ ~ ~ I ~.~~ ~. ! ~"'J\I' I'l'OI"
oono<
tu~O< oo('n< ot>QI) <0000 " 0
"I' ~~~~~~~ ~f~iig ~:O)'> 0 0000 0 ~!~ g~;:
J
0 uOJ)' OUJu 'lHW!,.' U ~uu • CI II:
(.
r
o·
j
OOOO')O)I)OO:)'OOOOt»O) ,0000,000c,~()Oo,00ooOl,ooo,0000,0000,oO()o'gro l~!g~ )O'UI~ "(l ~g~ggg!g'GOJ:lO OJUllO ~n~gggnggn~g~~~' :.J'.,'v '('(,1 (" ()CO(l·(onouorl >")ot!. (ll O(l'Ol,o(.('r;I'{'>"JOQ' oc..OJ' ooon. Qr)OO·,~oo~
,0000' oo~:> I :lO:l:l_')OllO_OJO) _0000_ 0000' DODO-DODO '000000000 .oooo.nooo·nr·! I flnn~
ru
000000'00»)'):1:10' lOuOoO)O::l _ 0000'0000' (l01l0'0c..00' 0000_0000' 0000.00000 ~~OO >~ ," l1.~::22~O
1'0000100:1:)' :1:):10. :)0 ... 0_0)01t _0000' oooo.OOl'O'OllVO."QOO' 0000
.0000.oon~:iL~!Sl~ ~n~~ ~ ~~
.::::11
Il~O
I '0' Ul'>O> U I 1(1'0 I ! g g ~~
gIg'
0000' 0033'0000_')01,1000,)00 'uOOO'OODO'OOOO'
))»0)00> J() :0< J ,,;). 'Jr.·),).O"Oo)
• ot" t >(lOOI' >rut u' DC [,0.1)1 "0' oOuo> OO{\O. rc t' ". G{ 00' r"luur >I ","),1. L1,('0. 0 10)' 00)0. ~:OO. JC "~", JJ 0 J'" ~),3 >a )00. )O~ n. 0 'J (,) >,~ ',1);"1. O{ O'J 1
H
,:~: gn~n gi~ hl~: ~~~ g ~ H~· CIn. UI ~u:;! i~; ~~u" • QuOl.' 0000' rl'{\lI .... n ' • 01'00' OOUO. oor C' l'w:;r. aoo~. Oll'lU' ["j~n >LU.' n. 0001'lI01 n >0300. JOlla. 0 ")'10' rn 10.0 }Ot>. Joro·
.:!!:::::::::::::::!:::::
~:~:~!
!f 1:
u La L'0 1
H~H~~U
h 0000' OO~~I:O J:lo')OOOo OlD) _ 0000' QOOthOO~O' 0'100.0(100.0000. OOOu' ~~~0>0 ~ ~ 0'0 al~r~t~~rt.~. III!bIl 8110 I">. ill'" • 0 I ) I • 0 I
It
'))lJJ' _
00)0' 0 )tl,l' gg~ ~ ~g ~::~ ~:~~: ~ : : g g n:lI • (lace' DODO> O(){lIl> OO('~. COOO. u rlOO> OuCO >('U!;C'>lI('on. oonrh ut ,~". 1101'0' 0000' u(')o. o')urJ' onuc.o 10J'lJ O'JO' 0)00 »0 ~O< O'h')< 0000· 00 01;.
•u I . , . 0 I ~o :t~:::~~:~:~~~~: 0 JOO' ~~~~: ~~~~. OCQ~· 0 I ~:;.O! 1~: ~~ U
ono.1) '"
i iIi ~ ~iI ~~ ~~~,
~('llO< ~OOO' o( 00' C00 J. v~01' 10 U) > U JJ' J"J ~ ~:~~ ~ A!~ ~ ~~~
':000< ':or U.lI .... (J( < 01''10. 001'0> (lOrD> r lIl't. (1/'".)1" OOlili. l nO/,. UOOO' O,'O'J' OOOC' (lODe. )OUO' OuIJu' t, J ) ) < O)iJ)' JUJ J. U J")' JOOO .ll/ Ql
:~~
H if
OUOIJ'VOOO'O::O'OO~~;~~~~:~l~~~~~t~~O'~~'l~~~ [o.~~~g~~~ ~ ~ OlOu.1t I ~ ... 0 I g~
!~~~ Z!~~ !~: ~~~~.
gg
~g~U~~ g;." I
j
J\. U!1
~:~ ~ ~l' o"uo • O~{ ('. 0000' N;CO' OI'I,L'>( (,00· rnoo. oorl' n~ J[)' 01
ini~~.
Or)
)')(J').
c..'l0.).
~
I) )'J_
>0,1"'1. ')Ql'U' C('"Q' "C'OO> oc 1'1·000,,, ')000. 0 )0)' VO ,:). O:J 0 () > ~u J J. ,j,/r,U >to '0 J." ~ uu 0
""uv • C 01,(,. ooor,< ClIUU> O( UO· 0.,00 >r{''ln. flo n ". tlU'.lQ.l){ ('0' OUIIO. C, C'." ,,1'0) OJ. (1)00 >lIl'O)' 0000' 10 vO' 0 IOU> r,n00 >0)0(0. JO{ 0> 'J'I"Q >01,10 ). C,C J[)'
·~ :~: :~:~ .:~:~ :~:~ :~: ~:~:[(! ;I~r~i!:i~:l ~ ~;i;"!i~: '~: ; i l ~JI !~i! !': :': 'i lili l\: :'!~'~: ;~ : :~:~: : : ~ ~: : : ': :~: : ~ : ':~,~: : ~: : ~:~ ~': : ~: : :;': ~: :
'100'0000.0000.0000' OOUO. 0000' 000'110000'OUOO' 00,)0.0000< 0000' 0000' UOOO. 0 )0:0' JOO OJ· ::nl o.,,~ JO. a 00" C~: 0> g g ; :>~~~. o~:HHr< 0 I : ~ t~~O!' O. ~,>. 0 I ~ Ln~ ~~ ~~~~ ~~e. 0 u 0.0 L~ l' J I ~O.::;:; ~ 1
-')oOOOO'OOOO.OOOO'OOOO'OUOO '000000000.9000'0000':0000'0000 'OOJO, OJor. ~OJO'O'lOl'IlOOO'OUOIJ,',ucr!>;.~:!;~
",:r
0 J lrFrU<:i~~' U.'lO·OI~O'OI )u.r Iv~,O I ~ "0 I ~ 1>0
-.0000. 0000
.
ooo~o. 0000' 0000 00000' OOUO oOvOO' 00 lO' 0:000 >UOOO> ~OC) 0 uOOO' ~ j~J 'Il~ ~o> ;'~~ i:~~L,; ~~t Ie> "
• '0000'0000 'OOJO ,0000'
t
t
~o 00' 0000 '0:) JO. 0000 '0000> 0.)00' 000)' OUOl' > ~u ,),j. g~ J " ~ g~~' ~ ~ ~t!~U:
r'
"'h 0000'0000. a 000 '00 OQ. 0)00' JOoo< 0000' 00 n< J ~~" >ooo~: i~U J ; ~ ~ ~: ~ ~ ~ v·" L"': ~!
......... 0000.0000. JO) »00 J ...
COMPUTERS and AUTOMATION for July, 1963
:::;;H !;"
H:::n:::::!::::!::: HH .
,~u; ~f ~~ ~ ~ ~ ~~ ~~~~ ~ ~ ~~i~ ~ ~ ~ ~; ~ ;,:!~! _.'
"ono·oooo. 0000. 0000 '0000' 0000' 0000' 0000' :1000. OJOJ '00 ):1. 0) OC <00)0. 0000. UOOO ·')uoo .~: :~>:lU J J • !r~ ~~ 101.' 0~P ). ~r,!.'" a ::::
L:
~~ ~
~ ~ g~;:::!:~ ~',~.
"; >,
,,~~;: ~ ,. ~ ; ~~;~: ~:
31
"ACROSS THE EDITOR'S DESK"
Computing and .Data Processing NewsleHer
TAHLE OF CONTENTS
New Applications
New Contracts
New Installations
Organization . News.
Computing Centers
Education News
32
33
34
35
35
36
New Products
Automation
People of Note.
Meeting News
Business News
~.lonthly Computer Census.
37
40
42
42
44
46
NEW APPLICATIONS
"LIVE COUPONS" READ BY
HONEYWELL SCANNER
Honeywell EDP, Welles ley Hills,
Mass., has reported that its optical symbol recognition system,
Orthoscanning, has achieved a
99.17 per cent document acceptance
rate in a "live coupon" processing
test. Other syst~ms which have
been tested on live coupons have
experienced rejection rates of
between 5 and 15 per cent.
Proctor and Gamble Company
provided the more than 125,000
coupons, which were scanned by the
system at a rate of 1370 coupons
per minute. The "live coupons"
(redeemed 'by c6nsumers and s~b
mitted by retailers for reimbursement) were considered completely
representative in terms of coupon
condition. Many were badly defaced from handling by the consum-
ers and retailers. Typical defacement included water soaking that
caused misshaping, spindling,
tearing, partial obliteration of
the codes, and holes in the documents. The coupons were standard
punched-card documents on which
was imprinted P&G data in Honeywell
Orthocode.
Orthoscanning reads series
of small vertical bars of varying
widths, called Orthocode, which
represent pertinent data. Conventional systems read letters
and numbers. Orthocode also contains "ortho-correction" information that permits automatic and
immediate regeneration of obliterated data. A Honeywell computer
in conjunction with the document
transport permits automatic regeneration of data at microsecond
speerls.
EDP IN OPERATION AT MASS.
REGISTRY OF MOTOR VEHICLES
-- Uadly-defaced live coupons
are examined by Honeywell EDP
engineer.
32
At the Massachusetts Registry
of Motor Vehicles, Boston, Mass.,
it is expected that taxpayers will
be saved some $6 million a year in
vehicle excise tax billing through
the use of a GE 225 computer system. Included in the total savings will be: $3 million in
revenue, which was formerly lost
when vehicle owners moved out of
the state ahead of the delivery
of vehicle excise tax bills; $1.5
million in clerical economies; and
$1.5 million in interest on loans
formerly paid out by cities and
towns. (Such loans were taken
out in anticipation of tax
revenues.) Earlier collections of
excise taxes will improve free cash
position of towns and cities by
some $15 million.
Under this new tax billing
system, two punch cards are prepared for each vehicle registered.
They are read into the computer at
a speed of 1000 cards/minute; the
computer verifies the data and
then writes it on to magnetic tape.
(One tape may contain as many as
70,000 registrations.) A programming method, developed by GE computer personnel, assigns a valuation to each vehicle according to
regulations of the Dept. of Corporations and Taxations. This valuation also goes on a magnetic tape
file, at a speed of 60,000 vehicles
an hour. At the rate of 18,000 an
hour, excise tax bills are prepared
and ready for mailing to vehicle
owners by cities and towns. Under
the old system, the time elapsing
between original registration and
receipt of the tax bill was about
fourteen months -- time now
required is about four months.
The new system will also be
used to: (I) maintain on magnetic
tapes registration records, police
listings of vehicles, and information on drivers' licenses; (2)
identify total vehicles; (3) assemble and analyze accident records
for safety programs; and later, (4)
to develop a master file of consolidated information connected with
each Massachusetts driver.
The GE-225 system includes a
central processor with an 8000
word memory; a 1000-card-perminute card reader; a 250-card-
COMPUTERS and AUTOMATION for July, 1963
•
Newsletter
per-minute card punch; two highspeed printers (900 lines-perminute); one output typewriter;
and eight tape units.
WORLD'S LARGEST, MAN-MADE
HARBOR GETS ASSIST FROM
COMPUTER
In planning the world's
largest, man-made port facility,
a three million dollar computer
system is being used at Computer
Sciences Corp., El Segundo, Calif.
The Univac 1107 computer at CSC
will help to fill a J-shaped
parcel of real estate extending
over two miles into the Pacific
Ocean from the present Long Beach
shore line. The present site of
Pier J lies 47 feet beneath the
surface of the Pacific. In the
construction of this pier, about
320 acres of ocean will be displaced by 3 million tons of rock
and 33 million cubic yards of
earth fill. CSC will use a General
Dynamics SC 4020 digital plotter
system to provide a graphic representation of the construction process as it is developed on the
computer.
In addition to its ocean filling computations, the computer will
handle a variety of management
planning functions including:
statistical evaluations of current
operations at the harbor; dollar
volume of various commodities;
duration of ship stays; number of
berths visited; and a daily census
of ships in port. Statistical
analysis of the reports will also
be used in the planning and control o~ ship traffic in the transition from present to new
facilities.
AUTOMATIC DRAWING OF
PERT CHARTS
Automatic drawing of the
charts which form the basis of
PERT has been successfully demonstrated by North American Aviation,
Inc., Los Angeles, Calif. PERT
(Program Evaluation and Review
Technique), a management planning
and control technique, was developed by the Navy in 1958 for the
Polaris program.
The extensive charts, or networks, which are the chief tool of
PEHT, show all activities which
take place, the events which must
be accomplished, and their interrelationships which lead to completion of a project. Adjustments
mus t be made while the project is
under way to keep the program
operating smoothly. Keeping the
networks up to date manually may
become almost impossible, because
of the complications.
North American Aviation, some
months ago, placed a pilot system
in operation which showed that
automatic updating of PERT networks was possible. The system,
called PERT-NAP for PERT Network
Automatic Plotting, makes it
possible to feed changes into a
computer on one day and have updated networks back the next morning. Automatic plotting of the
initial PERT network is also
possible.
The company has now put a
minimum operational PERT-NAP system into active functioning. By
the ~irst of next year, the company expects to complete a PERTNAP system that can be applied to
any of the widely diversified
aircraft, space, electronics,
nuclear and rocket propulsion
activities of the company's six
operating divisions.
CDC RECEIVES CONTRACT FOR
MORE THAN $2 MILLION
North American Aviation's
Space and Information Systems
Division, Downey, Calif., principal contractor for the Saturn
second stage, has ordered highspeed digital computers from Control Data Corp., Minneapolis, Minn.
The contract is expected to amount
to more than $2,200,000.
The computers (designated
Control Data 942s) will be used
for automatic booster vehicle
checkout at Seal Beach and the
Santa Susana Test Facilities,
Calif., the Mississippi Test Operations Center, and Cape CanaveraL
Fla., test and checkout facilities.
The contract provides for six computers, 24 tape units, two
printers and other equipment required to check out the Saturn
S-II launch vehicle.
NASA AWARDS CONTRACT TO
GENERAL DYNAMICS/ELECTRONICS
NEW CONTRACTS
PRIME SYSTEM CONTRACTOR FOR
THE RANGE SAFETY IMPACT
PREDICTOR ON ATLANTIC
MISSILE RANGE
Under a prime system contract,
the Air Force will lease two
CONTROL DATA 3600 Computer systems, from Control Data Corp.,
Minneapolis, Minn., for predicting impact points at the Atlantic
Missile Range. The real-time
computer systems associated input/
output equipment includes the new
CONTROL DATA 606 Magnetic Tape
Transports. The entire system is
valued at approximately $7 million.
It will be installed in a new
facility at Cape Canaveral with a
direct tie-in to Central Control
for Range Operations. Full operation of the system is not expected
until mid-1964.
RIVERSIDE TRUST COMPANY
BECOMES FOURTH BANKING
CLIENT OF NCA
The Riverside Trust Company
at Riverside, N.J. has contracted
with National Computer Analysts,
Inc., Princeton, N.J., to perform
its demand deposit accounting at
the NCA Datacenter in Princeton.
It is expected that NCA will begin the performance of this contract by the beginning of July.
COMPUTERS and AUTOMATION for July, 1963
The National Aeronautics and
Space Administration has awarded
a $60,1100 contract to General
Dynamics/Electrollies-Hoehester,
Hoehester, N.Y., for a study of a
ll1aUlleto-acoustie information storage system. As a result of research for more than a year, a new
technique for temporary or permanent storage of data is being used,
consisting of solid-state components, without moving parts. The
system is expected to provide
read-out rates of less than a
microsecond.
THIN-FILM DEPOSITION
TECHNIQUES UNDER STUDY
BY SYLVANIA
Sylvania Electric Products
Inc., Waltham, Mass., is conducting research on the properties of
active thin-film microelectronic
devices under a contract awarded
by the United States Air Force,
Aeronautical Systems Div., WrightPatterson Air Force Base, Dayton,
Ohio. The work is being performed
at the microelectronics laboratory
of Sylvania Electronic Systems.
Successful deposition of
thin-film devices, such as diodes
and transistors, on polycrystalline substrates would greatly reduce the production cost and increase the reliability of advanced
microcircuits through the elimination of many of the man-made connections required with conventional
methods.
Newsletter
DIGITAL COMPUTER FOR
OCEANOGRAPHIC DATA SYSTEM
CBS WILL USE TWO TRW-330's
FOR TV PROGRAM SWITCHING
Thompson Ramo Wooldridge
Inc., Canoga Park, Calif., has delivered a TRW-130 (AN/UYK-l) Digital Computer and peripheral
equipment to The Bissett-Berman
Corp., Santa Monica, Calif., for
use in a system for measuring,
recording, and transmitting oceanographic data. The computers will
be used as an integral part of a
shipboard oceanographic system.
The new CBS Broadcasting
Center in New York City, (scheduled
to go on the air in early 1964)
will use two TRW-330 control computers to aid in performing all
routine, preplanned program switching operations. Three distinct online functions which will be performed by the computer system are:
on-air continuity switching, facilities assignment switching, and
studio-lighting preset storage and
retrieval.
NEW INSTALLATIONS
GREAT BRITAIN LEASES
HIGH SPEED COMPUTER RECORDER
FOR ATOMIC ENERGY AUTHORITY
An S-C 4020 high speed computer recorder has been leased
from General Dynamics/Electronics.
San Diego, Calif., by the United
Kingdom Atomic Energy Authority,
Aldermaston, England. The S-C
4020 will record, onto 35mm microfilm or 9~ inch wide photorecording paper, the results of calculations on an IBM 7020 computer.
NASA TO USE CONTROL DATA
COMPUTERS FOR NIMBUS
WEATHER SATELLITE
The National Aeronautics and
Space Agency expects to install a
CONTROL DATA 924 Computer system
this month, for use in the Nimbus
Project for weather research. The
system will be utilized with an
identical 924 system currently in
operation. Each system consists
of two 924 computers, 10 magnetic
tape units and a variety of periphernl devices.
The computer systems will be
located at Command and Data Acquisition stations at Gilmore
Creek, Alaska, and Goddard Space
Flight Center, Greenbelt, Md. (An
additional station is being consider~d in eastern Canada.) At
the stations, the systems will be
used in gridding of picture data,
real-time assessment of the Nimbus
Satellite, and in long-term engineering evaluation of the satellite
and its subsystems.
UNIV. OF MICHIGAN BUYS
CONTROL DATA 160-A SYSTEM
Under a grant from the National Science Foundation, the
University of Michigan, Ann Arbor,
Mich., has bought a CONTROL DATA
160-A Computer System. The system,
installed at the University's
Meteorological Laboratory in June,
will be connected to a large-scale
analog computer by special interface equipment built by Control
Dnta. The purpose of the research
project, in which this hybrid system will be used, is to construct
a computer model of atmospheric
turbulence and diffusion. It will
also be used in other phases of
advanced weather research.
34
FIF.TH COMPUTER INSTALLED AT
BRITISH FINANCIAL EDP CENTER
The Financial Computing Center, St. Alphage House, London,
England, has installed a fifth
computer system, possibly making
this center the largest and most
completely equipped service of its
kind. The newest processor, a
National Cash Register 315 system,
supplements an NCR already in use
and three NCR-Elliott 003's. The
315 computers installed at the
center include NCR's CRAM electronic filing system (Card Random
Access Memory).
CANADA'S BELL TELEPHONE
USING HONEYWELL 400
The Bell Telephone Company of
Canada has installed a Honeywell
400 system at its Montreal office.
The system will be used to update,
once a day, available information
for all of its Yellow Pages directories. In addition, the Honeywell 400 will process Bell Telephone's stock transfer records and
aid in the assignment of dial
equipment in telephone exchanges
throughout the Bell territory.
The H-400 system includes a
central processor with 2048 words
of memory, 4 magnetic tape units
with a transfer rate of 96,000
decimal digits per second, a paper
tape reader, high-speed printer,
card reader and card punch.
COMPUTER INPUT/OUTPUT DATA
LINK DELIVERED TO MELPAR, INC.
-- Section of British financial EDP center shows NCRElliott 803 computer in background. High-speed printer
at right foreground, part of
an NCR 315 system, is preparing a financial report at
600 lines a minute.
The center specializes in
processing work for the British
financial community, although it
handles other commercial data
processing as well.
UNIVAC SYSTEM DELIVERED TO
U.S. ARMY CHIEF OF ENGINEERS
The office of the U.S. Army
Chief of Engineers, Washington,
D.C., has installed a UNIVAC 1004
Card Processor, at its data processing headquarters. The system
will help keep pace with the increased data processing required
in the office.
Adage, Inc., Cambridge, Mass.,
has delivered a computer input/
output data link to Melpar, Inc.,
for use in pattern recognition
research and information processing; This data link processes
analog signals for digital entry
into an IBM 1410 computer on a
real-time basis. Speech analysis
is one example of the use of this
system. Melpar, Inc., has been
actively engaged in speech research
for some time and expects the
Adage data link will open new
frontiers in this and other areas.
ANALOG COMPUTERS INSTALLED AT
HUGHES AIRCRAFT
Beckman Instruments, Inc.,
Richmond, Calif., has delivered
two analog computers to the Hughes'
Space Systems Division of Hughes
Aircraft Co., EI Segundo, Calif.
The computers will be used to
solve design problems in the con-
COMPUTERS and AUTOMATION for July, 1963
•
Newsletter
trol and guidance systems of the
Project Surveyor space vehicle and
to simulate landings of the space
craft on the moon.
TWO CONTROL DATA SYSTEMS
FOR AEC
t.
Last month Control Data
Corporation delivered the second
of two high-speed computer systems
to the U.S. Atomic Energy Commission for use by the Lawrence Radiation Laboratory, Livermore, Calif.
Both systems are standard production models. One system is the
CONTROL DATA 1604-A; the other is
the CONTROL DATA 3600 with a 64,000
word memory unit. The two systems
will be used by the laboratory,
together with seven existing computer systems, in the solution of
complex scientific problems.
ORGANIZATION NEWS
NEW ORGANIZATION CONCEPT
FOR PHILCO
Philco Corporation as outlined
plans for more fully using its
large-scale computer capabilities
in support of the expanding needs
of defense, space and other government agencies:
1. The present Computer Division will be consolidated with
the Communications and Electronics
Division. Total systems capabilities are expected to be further
strengthened by the consolidation.
•
2. Dr. S. Dean Wanlass,
Philco vice president and former
general manager of the Computer
Division will become Vice President-Technical Planning, with responsibility for the whol~ corporation for coordinating, planning
in computers, communications, command and control, and space
systems.
The new organizational concept is designed to intensify
Philco's efforts in the area of
military and related command and
control systems.
NAME CHANGE APPROVED FOR
REEVES
Stockholders at the annual
meeting of Reeves Soundcraft Corp.,
New York, approved a resolution to
change the name of the company to
Reeves Industries, Inc. Reeves
Industries, Inc., listed on the
American Stock Exchange, will continue the ticker symbol RSC. The
name change was made necessary
because of increasing diversification.
In addition to the name
change, a new corporate identification system is being developed
incorporating the use of a large
"R", which will be used by all
divisions as well as the parent
company. The Soundcraft division
will continue to use the established "Soundcraft" brand name on
all of its recording products.
PRINTED CIRCUIT PLANT ACQUIRED
BY CONTROL DATA
Control Data, Minneapolis,
Minn. computer manufacturer, has
acquired the assets and business
of Beck's, Inc., St. Paul, Minn.
printed circuit manufacturer.
The New York Stock Exchange has
approved 5823 shares of Control
Data common stock for listing in
exchange for all of the assets of
Beck's, Inc.
The new printed circuit subdivision will make components:
for military, spite(! :tnd indust.rial
electronics; for Cont.rol Ditta computers and peripheral equipment;
and to meet requirements for other
electronic manufacturers.
HONEYWELL COMPUTERS
MANUFACTURED; MARKETED IN
JAPAN
Honeywell data processing
systems are being manufactured
and marketed by the Nippon Electric Company in Japan. Six systems, Honeywell EDP's entire line,
are being marketed by the NEC
sales force. The company also
has begun production of its first
Honeywell system, the H-400.
Under terms of a licensing
agreement signed with Nippon Electric a year ago, the Japanese firm
will produce and sell Honeywell
computers under the Nippon Electric name. The systems are marketed as NEAC 2400, 3400, 2800,
and 3800 corresponding with the
H-400, 1400, 800, and 1800 nam~s
respectively. The recently-announced Honeywell 800-11 and
1800-11 systems will be called
NEAC 2800-11 and 3800-11.
Initially, Honeywell EDP will
supply NEC with many packages and
components used in the systems.
COMPUTERS and AUTOMATION for July, 1963
Later, Nippon Electric will manufacture its own components or obtain them from Japanese suppliers.
All software will be provided by
Honeywell on a continuing basis.
COMPUTING CENTERS
BUSINESS SERVICE BUREAU WITH
A DIFFERENT CONCEPT
A new service bureau, operated jointly by General Dynamics/
Electronics-San Diego, and Recordak
Corp., a subsidiary of Eastman
Kodak Co., Rochester, N.Y., has
been opened in Washington, D.C.
This facility is equipped to
translate computer output codes
directly into understandable language and drawings on microfilm
and paper. Computer centers in
Government and industry will be
able to have high-speed conversion
of computer "records" to a more
usuable form on an hourly or job
basis.
The heart of the new service
bureau is the General Dynamics'
eompul.er recorder, known as the
S-C ,1020. Th i s dev i ce t ra ns forms
maunetic-tape recordinus into usable language and graphics on both
page-size photo-recording paper
and on compact microfilm for in~tant accessibility and automated
information retrieval. Complex,
annotated, charts or drawings can
be recorded in less than a second.
Multiple film or paper copies are
produced from the primary microfilm records, using Recordak
equipment.
Magnetic tapes brought to the
S-C 4020 are electronically converted into combinations of printed characters, lines and curves by
use of a special cathode ray tube.
Directly opposite the tube screen
is a 16mm or :~;imm recording camera
and another optically aligned
35
Newsletter
camera for producing paper copies.
Business forms, maps, company symbols or. other fixed information
may be superimposed on the film or
paper by using automatic slide projection which is part of the basic
unit. The S-C 4020 is compatible
with existing automatic microfilm
storage systems.
Data from magnetic tape is accepted at rates up to 90,000 sixbit characters per second; printout
is at speeds in excess of 17,000
alphanumeric or symbolic characters per second.
: GE EXPANDS CHICAGO CENTER
General Electric Company has
installed a second GE 225 computer
system in its Information Processing Center, Chicago, Ill. The new
system makes the GE center one of
the largest in the Midwest. It
presently has a GE 225 and a GE
210 system.
The center handles: demand
deposit accounting for suburban
banks as well as Chicago's Cosmopolitan National Bank; payroll
processing, inventory control,
linear programming and Critical
Path Method scheduling, for local
customers.
COMPUTER CENTER TO. BE
ESTABLISHED AT AMERICAN
UNIVERSITY OF BEIRUT
with no limitation on the learning
time required or the number of
persons to take the course".
EDUCATION NEWS
TEACH COMPUTER PROGRAMMING
WITH NEW TECHNIQUE OF
PROGRAMMED LEARNING
The new method of instruction called "reinforced learning",
deriving from Prof. B. L. Skinner
of Harvard Univ., has been applied
to teach students how to program
a computer. The course is designed for the NCR 390 computer.
The technique makes use of
five psychological factors which
have been found to accelerate
learning:
(1) Information to be learned
is broken into small steps -the student absorbs only "one
piece" of information at a
time;
(2) Active response by the
student to each item or step is
encouraged, to stimulate interest and focus attention;
(3) Immediate response from
the machine is provided, to
"reinforce" correct answers
and "extinguish" incorrect ones;
(4) Se1.f-pacing is buil t into
the course;
(5) . A low-error rate is assured,
serving to build confidence and
keep interest high.
American University of Beirut
has announced plans .to establish
the first major academic computer
center in the Near East. The
center will be located on the
campus in Beirut, and will include
an IBM 1620 scientific computer,
an IBM 1410 data processing system,
and associated units of punch card
equipment.
The AUB, a privately-supported
institution, has an international
faculty of over 500 and a student
body of almost 3000 from some 50
countries. The center represents
a significant contribution to the
long range development of the
Near and Middle East.
-- Susan Horstman takes new
tape-recorded "reinforced
learning" computer program:ming course. It can be given
to a number of students simultaneously through earphones.
The workbook is for students'
responses.
NCR also is using a reinforcedlearning course to teach the operation of its new Class 450 bank
proof system. This course has cuT
in half the training time previously required and has increased the
confidence and ability of proof
machine trainees.
BELL TELEPHONE SYSTEM
BUSINESS COMMUNICATIONS
SEMINAR
The Bell Telephone System i~
90nducting an informational program for management, t~ assist
business executives to unaerstand
ways that modern communications
can contribute more effectively to
corporate planning, growth, and
profit. The facilities for the
seminars are at present located at
One East Wacker Drive, Chicago,
Ill.
The continuing seminars are
held in both one-day and two-day
sessions. The sessions are planned
for executives representing all
types and sizes of business, government and military agenc~es and
any type of organization that relies on business communications.
The seminars cover:
1. "The Information Explosion" -- an examination of the increasing volume of information
that must be organized and understood by management;
2. "The Integrated Management Information System"
3. "Communications" -- its
role in effective organizational
planning, growth, and profit,
with a practical explanation of
the technical nature of communications;
4. "Planning an Integrated
Communications System" -- with
attention to function, volume,
distribution, message characteristics, speed, accuracy, and cost;
5. "Building for Tomorrow"
-- covering products of today:s
research, and inter-relationship
of man and machines in the age of
the "information explosion".
The course will be offered
by NCR's branch offices for training employees of 390 users in
basic programming. The new approach will permit training of
the employees "when and where it
is most convenient for the user,
36
COMPUTERS and AUTOMATION for July, 1963
Newsletter
small size, low weight, modular
building blocks, parallel word
tran~fer and operation, two memory
sec-tlo~S, and two processing centers which rate independently.
There is an internally stored program and an instruction list of 75
commands. The random access core
memory of 8192 words is expandable
to 32,768 words.
NEW PRODUCTS
Digital
IBM 7094 II-IBM'S MOST
POWERFUL COMPUTER TO DATE
Monica is designed for use
in inertial guidance, flight control and automatic checkout as required for missiles, space vehicle~, advanced aircraft, and navigation systems for submarines.
One member of the family, a navigational computer, will be ready
late this year. Another, a cen~r~l com~uter with larger capabilIties, will be ready in the spring
of 1964.
IBM Corporation
Data Processing Division
White Plains, N.Y.
I~crease~ speed and processing
power IS provided in the IBM 7094
model II. Substantial increases'
in the internal processing speed
have been provided by a new instruction processing unit, which
reduces the number of cycles necessary for multiplication and divisi~n; and a new core storage unit,
which allows a higher degree of
overlapped operation in its 32,768
word memory. The system performs
a basic operation cycle in 1.4
millionths of a second -- 30 per
cent faster than the two-microsecond cycle of the 7094.
.
The higher degree of overlap
IS made possible by interleaved
ad~ressing in the new core storage
unit, providing in effect, two
separate memory banks, each with
a capacity of 16,384 words. Two
instructions may be retrieved
simultaneously or one retrieved
while the previous instruction is
being executed. A continuous flow
of overlapped instructions is
possible until the sequence of
instructions is terminated.
The 7094 II can execute, without change and at the higher speed,
programs now being used with the
709t1 and 7090 data processing
systems.
MONICA
MINIA TU~~ED COMPUTERS
Autonetics
3370 E. Anaheim Rd.
Anaheim, Calif.
Integrated solid-state circuits are being used in this company's new family of Monica microminiaturized computers. This division of North American Aviation
Inc. has developed the Monica lin~
through its experience in the manufacture of airborne military
computers.
Digital-Analog
II
~
SDS DIFFERENTIAL EQUATION
SOLVER
Scientific Data Systems, Inc.
0011 Alden Drive
Los Angeles tlO, Calif.
This eompany has announeed
the DES-I, a differential-equation
solv~r, which combines in a single
machine features of both digital
and analog computers. The DES-l
includes a central processor with
8192 words of overlapped memory,
~aper-tape reader and punch,
Input/output typewriter, console,
real-time clock, and graph
plotter. This gives it the problem-solving capacity of between
two and three fully expanded
analog computers.
The computer allows problems
to be changed in minutes, without
the tuning usually required in
analog computers. Magnetic tape
storage of problems and previous
results, permits a library that
minimizes programming effort.
The computer performs real-time
digital simulation while retaining the programming simplicity
of analog computers.
The company expects to deliver the first DE5-l in February of 1964.
Monica is a successor to the
Verdan and the 0-17 Minuteman realtime digital computers. It has
COMPUTERS and AUTOMATION for July, 1963
Software
AID FOR SIMULATION OF
INVENTORY SYSTEMS
The General Electric Computer
Department, Phoenix, Ariz., has
developed a programming aid, called
TRIM, for the simulation of inventory systems. The new software
package is an analytical tool used
by engineers in the design and implementation of improved inventory
control systems -- basically a
model which approximates the real
system. The TRIM approach assures
a workable system before it is
bu il t.
TRIM uses a GE-225 general
purpose computer, causing it to
behave like a complete, singlestage inventory system. It cons~sts of ~ix subroutines or operations, which are performed in
sequence. The routines process
customer demands, estimate future
requirements, place and receive
repl~nishment orders, purge overage Inventory, and cancel overextended back orders.
The new programming has at
least three significant features:
(l) i!. performs much fa s ter than
an operating system on the factory
floor, and it can simulate 50
time periods of activity in inven,tory systems in less than five
minutes;_ (2) TRIM can explore inven
tory situations without risking
monetary investment; (3) it is
possible to try all alternatives
and select the best method of
operation.
In a series of tests within
the company, TRIM has demonstrated
inventory reductions of up to 30
per cent. Customer service has
been given a lift, clerical costs
reduced, and risks involved in
designing new systems minimized.
TRIM will be made available to
current as well as future users
of the GE-225.
COBOL COMPILER FOR
HONEYWELL 400
Honeywell EDP, Wellesley,
Mass., has relased to users a
COBOL compiler for the H-400 data
processing system. It contains
all of the language elements of
COBOL '61 as required by CODASYL,
plus a set of elective features.
Electives implemented in the compiler include: segmentation
MOVE corresponding, the ENT~~
verb, the USE verb and a variety
Newsletter
of input-output techniques, such
as "DEMAND READ" and "DEMAND WRITE".
including money, to competing activities in more than one large
project simultaneously.
The compiler will operate under control of the Honeywell 400
monitor. Using Honeywell's program tape maintenance (PTM) software, object programs may be combined with EASY assembly system
programs to create a mixed COBOL
and EASY master program tape.
IBM 1401 CHANGES EASY WITH.
DATATROL PATCH GENERATOR
Major features of 400 COBOL
include ability to accept batched
source programs, automatic segmentation, and a competitive compile
time, which averages 20-25 minutes
for a typical 2048-word memory object program. The "compiler can be
used with any H-400 data processing system with a 2048-word memory
or larger, and a minimum of four
magnetic tape drives.
NEW COMPILER FOR
RPC-4000 OF GENERAL PRECISION
ACT IV, a new algebraic compiler and translator L is now available for the RPC-4000 from the
Commercial Computer Division of
Information Systems Group, General
Precision, Inc., Glendale, Calif.
The compiler is simple enough
to be used after a few days' training. The computer can be used with
the compiler without any knowledge
of machine-language programming.
Arithmetic and other sub-routines
required for running the translated program are included.
A new service program, available from Datatrol Corp., Silver
Spring, Md., enables an IBM 1401
programmer to easily make patch
corrections to an assembled 1401
program deck. The Datatrol Patch
Generator (DPG) technique uses the
1401 to generate its own patch
cards. With the DPG, the programmer compiles a list of corrections for punching in a format
similar to Autocoder or SPS. Then
the DPG is used to punch out any
required number of patch cards in
a single run. These patches contain up to four instructions per
card and are placed either permanently or temporarily in front oithe final transfer card of the
program deck and it is ready to
run.
The advantages claimed for
DPG are savings in programming
time by reduction of time lost
through clerical error, and savings in computer time spent on reassembly runs. A condensed operational program and instructions
for using the DPG is available
from this company.
Data Transmitters
and AIC Converters
POCKET-SIZED RECORDING DEVICE FOR USE WITH IBM 1030
IBM Corporation
Data Processing Division
White Plains, N.Y.
A pocket-size recording device has been developed by this
company which now permits factory
employees to enter production information 1nto an electronic datagathering system. The new IBM
1030 data collection system
gathers information from electronic
in-plant reporting stations and
transmits it directly to a computer for immediate analysis. The
computer may be located in the
same building or thousands of
miles away.
A programming manual for the
compiler defines the vocabulary
and rules of language and includes
operating instructions, error location and correction methods,
sample programs, internal number
representation for the computer,
and reference lists and tables.
RAMPS FROM C-E-I-R
RAMPS (for Resources Allocation and Multi-Project Scheduling)
is a software management tool which
reduces costs and centralizes decision-making for business and industry. It has been introduced by
C-E-I-R, Inc., Washington, D.C.
RAMPS helps managers to decide in
~avance when and where men, machines, materials, and funds should
he assigned, and to predict how
much time will be required by each
undertaking. Using IBM 7090 or
IUM 7094 equipment with RAMPS,
business and industrial management may allocate limited resources,
38
-- An employee at his work
location logs information on
the plastic cartridge.
He stores the cartridge
in his pocket.
-- When he is ready to send
information to the computer
he can insert it in an electronic reporting station.
COMPUTERS and AUTOMATION for July, 1963
Newsletter
Memories
NEW MEMORY-STORAGE DEVICES
•
Information Systems Group
General Precision Inc.
Glendale, Calif.
.
Some new random-access storage
units have been developed and are
being marketed by this company.
The fast-access, rotating memorystorage devices include three magnetic-drum and ten magnetic-disc
models.
The Series LIOO discs use one
side for storage, while the L200
discs use both sides. Flying
read/record heads are fitted to
both. A read/record head is used
for each track, eliminating the
need for head movement. When the
disc rotates, the flying heads
float over the surface on a very
thin cushion of air, automatically
compensating for any variations
caused by temperature or other
factors. Sizes of the discs range
from 4 to 24 inchei in diameter.
Disc-storage capacity ranges to a
maximum of 7,680,000 bits.
-- A technician at the Information Systems Group of General
Precision, Inc., test a readwrite head that records and
withdraws information from
magnetically coated surface of
a giant, 48-inch-diameter magnetic-memory disc. The disc
is part of a six-disc memory
system that can store up to
IS3 million bits of information and can retrieve information on the basis of content
alone (rather than location
of information) in 33 thousandths of a second. The memory system could be auxiliary
to a computer or data-processing system.
The L500 series of magnetic
diums range in capacity from
512,000 to 4,100,000 bits. Each
of the three models is available
in two types of different lengths.
All models have an integral aircirculation system, and high-precision, lifetime-lubricated bearing assemblies.
The new memory is being restricted to STRETCH where it can
be studied further in an operating system. IBM does not plan to
sell it in its present form in
commercially available computers.
Both of the new random-access
storage units are now available
for extending the capacity of computer systems used in commercial,
industrial, and military
applica tions.
300-NANOSECOND CYCLE TIME
MEMORY
FIRST TUNNEL-DIODE MEMORY UNIT
IS INSTALLED IN
OPERATING COMPUTER
IBM Corporation
Data Systems Division
Poughkeepsie, N.Y.
A tunnel-diode memory unit
has been installed in an operating
computer by this company. The
tunnel diode, one of the fastest
switching devices known today, has
SWitching-times measured in billionths of a second. A number of
tunnel-diode circuits and devices
have been proposed before, and
some models have been built, but
the new memory is the first repur ted 1.0 bl! llsed in i111 opera I. i IIU
system.
The new memory is a small,
high-speed register unit installed
as a specially engineered feature
in the IBM STRETCH computer used
at the company's Poughkeepsie development laboratories. Its purpose is to modify instructions to
the computer at extremely high
speeds. In the STRETCH system,
the memory fetches and stores data
in a cycle time of 600 nanoseconds
(billionths of a second). In engineering test, cycle times have
reached 200 nanoseconds -- at
least three times as fast as needed by any computer, including the
STRETCH.
Its full capacity is i f computer words of 74 bits each. At
the 200 nanosecond cycle, the
memory can process over 45 million
letters or digits a second -equi valen t to 90 full-length novels.
The basic component in the
IBM register memory is a tiny
circuit containing a tunnel diode,
a resistor and an inductor, all
encased in a plastic cell. A
total of 1250 of these cells are
mounted on two 4~ x 16-inch
printed-circuit cards, which are
plugged directly into the computer.
COMPUTERS and AUTOMATION for July, 1963
Fabri-Tek Inc.
Minneapolis, Minn.
The FFM-202, a self contained
magnetic film memory system, with
a 300-nanosecond total cycle time
is commercially available from
this company. Read or write only
time is 200 nanoseconds. Access
time is 150 nanoseconds. Four
operating modes are available:
read only, write only, read restore, and read-modify-write.
It is a complete memory pack~
age, with its own power supply
and will operate from any 115volt, 60-cycle ac line.
Magnetic film memory elements
are about 1000 angstroms thick and
are packaged in multilayer laminated overlays containing all the
drive and sense lines. A typical
8" x 10" memory plane contains
128 words of 36 bits each. Several of these planes connected
together will permit a variety of
memory sizes and word lengths.
Input-Output
ELECTRONIC 'ARTIST'
DRAWS STRAIGHT OR CURVED LINES
Benson-Lehner Corp.
Van Nuys, Calif.
This company has developed an
electronic "Artist", called Electroplotter II, capable of drawing
straight or curved lines automatically in UrCiphs, charts or maps,
in any combillCitioll of' four colors,
and annotCitillu them wit.h words or
Newsletter
symbols. The optional four-pen
plotting head which permits drawing in four colors also permits
four line,widths. The full alphanumeric printing allows complete
graphs to be drawn without subsequent hand operations. Electroplotter II accommodates graphs as
small as 8~ x 11 inches and charts
as large as 42 x 58 inches. No
special operating skills are
required.
SPEEDS UP TO 100-INCHES/SECOND
FOR NEW TAPE REELERS
Components
MAGNETOSTRICTIVE DELAY LINES
NOW PRACTICAL FOR
AEROSPACE APPLICATIONS
Digital Devices Inc.
212 Michael Drive
Syosset, L.I., N.Y.
This company has recently
introduced a new packaging concept
making aerospace applications of
magnetostrictive delay lines
practical. The new technique
uses solid castings of appropriate materials using circular contact areas to reduce stress concentrations and allow the use of
standard O-ring type seals.
The device can receive its
design and color instructions
from magnetic tape, punched cards
or tape, or directly from a computer. It will operate under program control or operator control.
Output FORTRAN subroutines are
supplied with each plotter to aid
in systems integration.
The plotter is designed for
use in data processing centers,
for graphing scientific, engineering, and economic data. A very
wide range'of applications is
possible.
NEW, FOUR-ROW TELETYPEWRITERS
Bell Telephone System
Avenues of the Americas
New York 20, N.Y.
Teletypewriters with four-row
keyboards and a 66-percent increase
in speed on regular teletypewriter
exchange service are now available
from this company in two models.
Both operate at 100 words per
minute using a newly developed
eight-1ev~1 Data Interchange Code
which provides 128 possible combinations for the formation of
characters and eliminates many
present information "interchange"
problems. Both models also are
equipped with dials for faster
station calling without need for
an operator. Speaker-phone units
permit "hands-free" monitoring of
call progress tones.
40
Provisions have been made in
the new design for evacuating or
purging with inert gases.
MUL TIAPERTURE CORE USED IN
NEW MAGNETIC LOGIC CIRCUITS
Omnitronics, Inc.
511 N. Broad St.
Philadelphia 23, Pa.
This company has introduced a
new series of tape reelers which
permit continuous bidirectional or
unidirectional operation of papertape equipment at tape speeds up
to 100 inches per second. At ten
characters per inch for conventional punched tape, this gives
1000 characters per second.
Double dancer arms on the
new OMNI-DATA Model RS-500 (unidirectional with rewind) and Model
RS-600 (bidirectional) reelers
allow very high paper speeds without excessive back-and-forth motion of the arms. The new series
is compatible with all OMNI-DATA
readers and recorders and with all
other paper-tape equipment. Both
reelers normally accommodate NAB
reels up to 8-inches in diameter.
Lockheed Electronics Company
6201 East Randolph St.
Los Angeles 22, Calif.
A new ferrite multiaperture
device has been developed by this
company for application in magnetic logic circuits. The new
component, called MAD-5 wafers,
is used in a line of magnetic
logic modules offered for use as
building blocks to construct computing and other digital equipment.
The MAD-5 devices, small 5hole ferrite cores, are wired to
form logic modules such as OR
gates, AND gates, and other circuits assembled in easy-to-handle
wafers. Encapsulated assemblies
also are available for resistance
to shock and vibr,a tion.
AUTOMATION
TRANSLATION FROM CHINESE TO
ENGLISH DEMONSTRATED AT IBM
An experimental system for
automatic machine translation of
Chinese to English recently was
demonstrated by IBM Corp. The
system is being developed at the
Thomas J. Watson Research Center,
Yorktown, N.Y., ijnder a contract
from Rome Air Development Center,
Air Force Systems Command.
The demonstration was made
with a limited vocabulary. A
large Chinese-English "machine
dictionary" is now in preparation
which will permit translations of
more extensive Chinese texts in
the future.
COMPUTERS and AUTOMATION for July, 1963
•
Newsletter
..
The system uses a special
input coding system and a large
capacity storage unit with associated processing circuits. The input system enables a non-Chinesespeaking operator to encode 6500
different Chinese characters. Its
capacity can easily be expanded
to 16,000 characters or more if
desired. The technique for classifying Chinese characters is presently being simulated by hand
coding but will be embodied in a
device under development.
.tfr.t ~t jjLbii ~.. i1'3 rtt' ~ at re~
OfJ1.j-. ,~1l ~ , ~~ ~t If] 't.,t -fiJ(
~~l!. 'P3 -t~1~ ~ J .
Word-for-Word Translation:
The prototype for the translating system is based upon the
Russian-to-English translation
system previously developed at
IBM. The major requirement in
modifying the system to process
Chinese is to build up a sufficiently large dictionary and to
specify linguistic rules precisely enough so that an accurate and
readaole translation can be
produced.
Recently discover/discovery magnetic core (de) switching
time possible shorten, therefore use/consume it come make
even high speed (de) storage device (Ie).
Machine Translation:
Recently discover switching time of magnetic core possible
shorten, therefore possible use it in order to make storage
device of even higher speed.
The input device being developed for the IBM Chinese language processing system is based
upon a classification of Chinese
characters by Dr. Lin Yutang,
scholar and philosopher. The device uses a regular typewriter
keyboard, but three keys must be
struck one after the other, to
determine one Chinese character.
Output of the typewriter is
punched paper tape which is fed
into the translator.
The input device is being
built jointly by IBM and the
Mergenthaler Linotype Company for
the Air Force Information Processing Laboratory. The linguists
and engineers at the IBM Research
Center are under the direction of
Mr. IIsien-Wu Chang. Mr. Fang-Yu
Wang, Yale University Institute
for Far Eastern Languages, is serving as consultant on the p~oj~ct.
Human Translation:
It has been discovered recently that the switching time of
magnetic cores can be shortened. They, therefore, can be
used to make storage devices of even higher speed.
-- Above is an example of a Chinese sentence and translations made
(1) word-for-word, (2) by machine, and (3) by a human translator.
The words "de" and "Ie" have no direct English equivalents; they
serve purely grammatical purposes in the sentence.
[gJ '~u ~20 ~'8 ~17 ~'. ~) ~25 ~ 08
11"'
2'
.=J:.
22
,;.g
fJ
T
-
r 28 ~26 ~.
m
~
•
a
0'32
;.
NEW RAIL CAR LOCATION AND
INQUIRY SYSTEM
A computer inquiry system has
been developed by Honeywell EDP,
Wellesley, Mass., which permits
railroads to verify the location
of rail cars in transit. The system links a high-speed digital
computer with the Bell Systems'
nationwide teletypewriter exchange
(TWX). Inquiries may originate at
any of 60,000 TWX terminals in the
U.S. and can be sent directly to
a central computer. The computer
processes, in seconds, a variety
of information on rail car status,
and automatically transmits the
answer to the sending terminal.
It.
P
S
:0
12
J.
)JJ.
[6J'O
-h
I..
A
?
,,'
tJ
2'
.A.
,<'-1,......
__
g, -....
~
-
-
~8 [2J'~ L3J4· ~""I
~27
~29
I
L!..I
)'1
Ju
J
7
~
P
1)1
RETURN
B
CAP
ENGLISH
MODE
~.'
~' ~!9
~ ~31
;;- 30 ~
it.. 6 ~
'..• 4 [lli2
e
p, 3[1fd~
i7
< 1" ~ [
! ]
.
" .,OJ JJ ,~ -\0) oJ.
}-+t
~
.
fvlr771 rt7l ~ ~ ~ p;)9
LJLJLJ~~~~
1
DOlT!
I
•
~
CHINESE
MODE
-- Above is the keyboard layout of the input .device being developed
for automatic Chinese translation. The marks on the keys correspond
to portions of Chinese characters. The operator recognizes one of
these characteristic marks at the top of the character and another
at its bottom. When the two keys corresponding to these marks are
depressed, the small group of characters containing both of them
will be displayed on a screen in front of the operator. The operator
will look at the characters and then press a third key specifying
which of the displayed characters matches the desired one. The input device will then punch a code on paper tape for t.he partieular
Chinese character.
COMPUTERS and AUTOMATION for July, 1963
Newsletter
Information on specific car
location, name of shipper and consignee, type of car, its owner and
car number, and the commodities it
is carrying, are maintained on a
master storage file in the computer.
Any of this information can be used
to originate an inquiry. (Current
rail inquiry systems require knowledge of the car numbers before additional information can be obtained.) The computer system can
also be used by railroads to maintain records of the location and
status of their rolling stock.
In a recent demonstration at
the Terminal Railroad Association,
St. Louis, Mo., the time from
inquiry to response for typical
cases was about four seconds.
The Terminal Railroad Association is a switching railroad
jointly owned by 15 of the nation's
railroads. TRRA maintains seven
major classification yards in the
St. Louis area, through which approximately one million loaded
freight cars are handled each year.
TRRA is participating with Honeywell EDP in a study of the application of computers and data communications in rail car accounting
systems.
PEOPLE OF NOTE
EDP CHAIRMAN ELECTED BY NOMA
Malcolm D. Smith, manager of
the special programming projects
department at Honeywell EDP, has
been elected national committee
chairman, data
processing committee, of the
National Office
~ Management Asso'ciation. As
chairman, Mr.
Smi th wi 11 be
responsible for
disseminating,
to chapters and
membership, the
latest information on automatic and EDP techniques. He also will supervise
arrangements for data processing
seminars, clinics and publications
for domestic and international
memhership.
DIRECTOR OF SCIENTIFIC
COMPUTING
Dr. Louis Robinson has been
promoted to director of scientific
42
computing for the Data Processing
Division of IBM Corp.
Dr. Robinson will be responsible for developing and implementing new marketing programs for
products and
applica tions in
the scientific
computing area.
He will coordinate exchange of
information
among IBM and
its customers
in advanced
computing areas and also will
supervise recruitment of graduate
scientists for work on computing
and programming systems for advanced scientific use.
_TELEREGISTER NAMES
EXECUTIVE VICE PRESIDENT
Luther A. Harr, Jr., has been
named executive vice president of
The Teleregister Corp., Stamford,
Conn. He has
been assistant
to the chairman, John E.
Parker, since
January. Mr.
Harr had previously been
director of
Univac operations for
Europe, Africa,
and the Middle East for Sperry
Rand International Corp~, Lausann~
Switzerland.
HUGHES COMPUTER DIVISION
HEADED BY ALTERMAN
Francis J.
Alterman has
been appointed
manager of the
computer division of Hughes
Aircraft Company's ground
systems group.
Before joining
Hughes, Mr.
Alterman was
president of
Advanced Scientific Instruments, Inc., in
Minneapolis.
The computer
currently engaged
ment of real-time
tems for military
application.
division is
in the developcomputer sysand commerical
MEETING NEWS
SJCC EX POST FACTO PROFILE
What?
Spring Joint Computer
Conference -Where?
Cobo Hall in Detroit, Mich.
When?
May 21-23, 1963
Who was there? Over 2400 registrants, and 3000 visitors to the
exhibi tion hall
What did they hear?
• Ray Eppert, president of the
Burroughs Corporation, said in
his keynote address that "electronic computation and information processing, supported by
management and intelligently
used, can and will write an effective economic accident
policy.
"Management must show th-at
it is flexible and can adapt to
changing times. New skills and
abilities must be mastered to
meet the spiralling demands of
our expanding economy~
In
fact, business needs a weapons
system concept, and computers
are vital to such a program."
• Walter A. Rosenblith, Professor of Communications Biophysics
at MIT, told the conference
luncheon that "what is needed is
not competition or coexistence
between brains and computers.
What is needed is skilled coopera~ion between the~ardware
and software of the comp.uter expert, and the 'pinkwar~ of the
physiologist to effect a fruitful marriage of the disciplines
so that a better understanding
will be obtained of both man and
his tools."
• Also: a sequence of 37 prepared papers, and four panel discussions on computer programming,
spacecraft simulation, data acquisition and display, etc., including the award-winning paper
by Douglas T. Ross and J. E.
Rodriguez of the MIT ComputerAided Design Project on the
theoretical foundation of the
computer-aided design system.
What did they see?
~Over $12.5 million in computers,
peripheral equipment, and services, displayed in 160 booths by
82 exhibitors.
• Notable among the exhibits
was Burroughs' first public dem-
COMPUTERS and AUTOMATION for July, 1963
•
Newsletter
onstration of the B5000. The
unit showed its capability to
process two programs simultaneously. Control Data Corp.
showed a new 1200 card per minute
reader, and its CDC 818 disc file
system. Digital Equipment introduced its new low cost computer,
the PDP-5 principally for applications in process control. DEC
also announced a PDP-6 computer,
medium scale, designed to control
experiments, collect, and analyze
data. Western Union announced
an optical character reader which
reads typewritten material in any
font and converts it into telegraphic code in the form of fiveunit punched tape for immediate
transmission. AddressographMultigraph disclosed details of
its new high speed printerprocessor.
Where can I get a copy of the
Proceedings?
Address mail order requests to
Spartan Books, 301 North Charles,
Baltimore 1, Md. Members of
AFIPS .... $5
Nonmembers .... $10
PRINCIPALS OPEN SJCC MEETING -- E. Calvin Johnson, general chairman of
the 1963 SJCC prepares to cut the opening tape as Brian W. Pollard,
chairman of the technical program, Willis Ware, chairman of the
Governing Board of AFIPS, Ray Eppert, president of the Burroughs
Corp., and J. Don Madden, president-elect of AFIPS, look on.
BEMA ACQUIRING SOME REAL MUSCLE,
CHAIRMAN TELLS SPRING MEETING; STANDARDS STRESSED
R. Stanley Laing, ch;:lirman of
the Business Equipment Manufacturers Association (BEMA), told about
100 attendees at the Spring Meeting,
May 0-10, at the Seaview Country
Club, near Absecond, N.J., "that a
great deal has been done in a short
time, and we now have a trade association with some real muscle,
able to assume responsibility in
representing this industry."
•
"For example, we have formulated some very specific plans with
respect to our standards effort,
both in this country and internationally. It is no secret that one
of the principal reasons for the
reorganization of this Association
(formerly OEMI, Office Equipment
Manufacturers Institute) was the
necessity for us as manufacturers
to exercise more aggressive leadership in creating workable standards for this industry."
Mr. Robert G. Chollar, chairman of BEMA's Data Processing
Group's Advisory Committee on Plans
and Po licy, told the meeting tha t
"I am convinced that progress in
the overall standardization program has been considerable." However, Mr. Chollar warned that the
interrelating responsibilities of
the standards program demand extreme care and caution to insure
tha t. all economic and technical implications of a standards adoption
are understood before final action.
"On the other hand", he said, "this
caution must be coupled with expediency ill the rapidly changing
field wherl! lack of standard practice is and will continue to stifle
growth and economic advance on
national and international levels."
In his talk, Mr. Chollar estimated that in 1962, 12,750 manhours were spent by BEMA member
company technicians in the standards program, representing an investment of $2,400,000 and the
time of some 700 individuals. Mr.
Chollar commented on this with the
paraphrase: "Think not of what
the standards program will do for
you: think of wha t the standards
program will do to you if you
don't participate."
"For years people have used
simple equations to describe the
physical universe, the part that
was the most easily understood.
Nowadays people are building more
complex mathematical models to
illustrate some of the more complex
activities. Simulators are actu-
At the luncheon talk on the
closing day of the meeting, Dr.
George W. Petrie III, Program
Director of the IBM Executive
School, remarked that the rate of
technological change in the business equipment industry recalled
the Danish proverb: "Prediction
is difficult, particularly when
dealing with the future."
Dr. Petrie commented that
"management science offers a great
many additional tools which management can use to sharpen its
decision-making capabilities.
Along with management science, a
person very often thinks in terms
of mathematical model building.
Actually this is a simple kind of
concept.
COMPUTERS and AUTOMATION for July, 1963
ally being built that will allow
management to take a broad look at
the entire activities of the firm
and see what the impact of reorganization will be on that firm.
The horizon sholVs us vastly increased use of simulation and vastly more powerful types of simulators
to be adapted to the needs of the
firm."
Newsletter
BUSINESS NEWS
BOXSCORE OF SALES AND PROFITS FOR COMPUTER FIELD FIRMS
C&A presents below comparative operating results
for firms of interest to comptl~er people" as distilled from the latest group of news releases.
COMPANY
PERIOD
SALES
Current Eeriod
Previous period
~48 2811 2000
$45,957,000
(+6.0%)
~6!472!901
(+11.5%)
Add ressographMu1tigraph
Three months
ending April
30, 1963
Ane1ex
Six months
ending March
31, 1963
$5,805,470
Three months
ending March
31, 1963
$98,967,000
Burroughs
Collins Radio
Control Data
Nine months
ending May 3,
1963
Nine months
ending March
31, 1963
~90;1602000
~117,356,000
~3824352937
Electronic Associates, Inc.
Three months
ending March
31, 1963
General Precision Equipment
Three months
ending March
31,.,1963
Honeywell
Three months
ending March
31, 1963
$l35,145,562
Three months
ending March
31, 1963
$453,226,278
Three months
ending March
31, 1963
$241,964,000
Six months
ending March
31, 1963
$20,060,000
Six months
ending March
31,1963
$23,776,000
ITT
ITEK -
Plicka rd Bell
Electronics
44
~ 1 2179 2000 '
$1,234,000*
~4,328!460
(-9.8%)
(-18.8%)
~147!5822300
~486!6572085
~27028682000
~16!054!000
~27!553!000
(+2.0%)
~2872661
(-6.8%)
~1, 722 2000
$2,120,000
(-18.0%)
~224082000
(+19.5%)
President Anderson sai d
that high R&D write-of f
caused the decline in
earnings.
President Eppert said
reduced revenues was
"due entirely ••• (to
the) ~hasing out of
certain defense programs "-.
$2,012,000
' (+39.0%)
~127762632
(+69.0%)
$1,055,146
(-4.9%)
(+50.0%)
$2,874,071
$48,200,000
$48,300,000
~329042000
NOTES
(%)
$308,320
$27,665,535
Three months
ending March
30, 1963
PROFITS
Current Eeriod
Previous ~eriod
$3,829,000
$144,607,000
Data Products
Corp.
IBM
(%)
$6 2600
-$24,000 (Loss)
~350,780
Figures do not reflect
recent acquisition of
the Bendix Computer
division.
*previous quarter.
Firm is just comp1etin g
its first fiscal year.
(+40.1%)
$249,570
(-0.5%)
~889!000
(+2.5%)
$866,000
(+8.9%)'
~621642156
(+16.5%).
President Binger said
"Improvement in profit s
••• due to EDP division
where good progress is
being made".
(+12.1%)
At annual meeting,
Chairman T. J. Watson,
Jr. , noted that IBM ha s
grown an average of 13 %
per year since 1914.
(+15.0%)
President H. S. Geneen
said that sales in the
United States increase d
by 22% du ring the peri ode
(+450.0%)
President Lindsay said
that R&D is concentrat ing
in graphic data acquis ition and textual data
processing.
$5,292,909
(+7.4%)
~63,0872363
$56,264,982
(+12.0%)
~924082000
$8,182,000
(-20.0%)
~314,000
$57,000
(+16.0%)
~752!000
(+12.6%)
$662,000
COMPUTERS and AUTOMATION for July, 1963
Newsletter
COMPANY
PERIOD
Sperry Rand,
Corp.
Fiscal year
ending March
31, 1963
Thompson Ramo
Woo ld ridge,
Inc.
Three months
ending March
31, 1963
SALES
Current Eeriod
Previous period
$1,227,085,610
$1,182,554,230
$117,351 2°36
$111,670,230
IMPORTS OF COMPUTERS
INCREASED 69% IN 1962;
EXPORTS UP ONLY 24% IN
PERIOD
A recent U.S. Department of
Commerce bulletin showed that imports of electronic computers have
risen by 69% in 1962 to a dollar
volume of $7,966,716 over $4,728,571 in 1961. According to the report issued by the Department's
Scientific, Photographic a~d Business Equipment Division, the increase came in the face of an
overall decline in the value of
imports of business equipment to
the U.S. rif6.6% in 1962.
Computer imports climbed from
4.9% of the total value of business
equipment imports in 1961 to 8.9%
in 1962. Canada is the largest
single supplier of computers to
this country, contributing $3,515,
545 in equipment or 44% of total
computer imports.
Exports of electronic computers from the U.S. rose 23.6% in
1962 to a dollar volume of $136,172,335 over $110,129,404 in 1961.
In 1962, computers represented
42.1% of the total exports of business machines from the U.S., a considerably higher percentage than
in imports. Exports of punch card
calculating machines declined 13%
in the same period from a dollar
volume of $58 million to $50 million. Overall, exports of business
machine&, including typewriters,
dictating machines, cash registers,
etc., increased 4.~~ in 1962 over
1961.
RECOMP LINE
DROPPED BY AUTONETICS
The Autonetics Division of
North American Aviation, Inc., has
announced that it will discontinue
the manufacture of its RECOMP line
of computers.
(%)
(+3.8%)
PROFITS
Current Eeriod
Previous period
~1323842794
NOTES
(-46.0%)
President Vickers sited
substantial losses in
the electronic data
processing and office
equipment divisions as
r~asons for the sharp
decline in profits.
(+7.7%)
President H. ft..;- Shepard said increased
shipments of computers
for military and commercial applications
caused revenue gains.
$24,373,816
(+5.1%)
$3,09°2 749
$2,878,166
"Although Autonetics is going
out of the commercial RECOMP computer business it will of course
continue its work on computers for
military and space use which has
constituted the great bulk of our
computer activity," President John
R. Moore said.
with a major electronics company.
Boston-based Raytheon has been
mentioned as a likely recipient
of the product line of IPC, since
'the large military electronics
house has been actively seeking
an "in" in the commercial computer
display and data communications
equipment field. The departure
of co-founder S. Paul Blumenthal
"RECOMP was des igned for a
from IPC recently helped fan the
small, specific market of indusflames-of the merger rumor. Howtrial and scientific applications.
ever, Ladd told C&A that no merger
It is a management decision not
to continue in this field."
or sale of the company is.be!!!.ll._
given serious consideration at the
current time. Also, he said, no
RECOMPS in inventory will be
offered fOT s'ale,"the company sard.' "inajor' change in' the"prodlfct line
or in personnel are expected.
Trained service personnel will be
absorbed in Autonetics' Computer
and Data Systems Division, in
order to continue to provide
CONTROL DATA CORP. ACQUIRES
service for RECOMP users.
CONTROL OF DUTCH FIRM
IPC UNDERGOES REFINANCING
Information Products, Inc.,
of Cambridge, Mass., has recently
joined the rather populous "no
cash" club of the computer field.
The membership ,requirement is a
propensity to lease rather than
sell one's products. Members
frequently are noted to engage in
such recreations as merging, selling-out, and filing bankruptcy
petitions.
The ~ year old Cambridge
maker of computer display equipment, coding keyboards, and symbol
generators is seeking another out
from non-exclusive "no cash" club,
according to Hammond Ladd, president. By July 1, the firm is expected to complete refinancing
with a major investment company~
During the period of negotiations,
Paul Atwood is serving as administrator of the company.
Ladd denied rumors that the
firm is contemplating a merger
COMPUTERS and AUTOMATION for July, 1963
Control Data Corp. has announced purchase of a majority of
the outstanding shares of Electrofact N.V. of Amersfoort, The
Netherlands, in exchange for
39,950 shares of CDC's common stock.
The Dutch firm manufatures
and markets a broad line of measuring, recording, and control devices and systems for use in industrial processes. However, its
international marketing and service
organization for marketing automatic control equipment in Europe
is the primary reason for CDC's
stock purchase, industry observers
believe. Electrofact offers CDC
a strong, established sales force
in the Common Market.
Electrofact will continue its
industrial control activities, but
it will also provide facilities
for the manufacture of CDC's computing equipment in The Netherlands. This is the first manufacturing fucility for CDC within
the Common Market. area.
1},5
C,ENSUS
MONTHL:Y
The number of electronic computers installed, or
in production at anyone time has been increasing at
a bewildering pace in the past several years. New
vendors have come into the computer market, and
familiar machines have gone out of production. Some
new machines have been received with open arms by
users -- others have been given the cold shoulder.
To aid our readeis in keeping up with this mushrooming activity, the editors of COMPUTERS AND AUTOMATION present this monthly report on the number of
American-made general purpose computers installed or
on order as of the precedi~g month. We update this
computer census monthly, so that it will serve as a
"box-score" of progress for readers interested in
following the growth of the American computer industry.
Most of the figures are verified by the respective manufacturers. In cases where this is not so,
estimates are made based upon information in the
reference files of COMPUTERS AND AUTOMATION. The
figures are then reviewed by a group of computer
industry cognoscenti.
Any additions, or corrections, from informed
readers will be welcomed.
AS OF JUNE 10, 1963
NAME OF
MANUFACTURER
.....Addressograph-Mulqgraph
Corporation
Advanced Scientific
Instruments
Autonetics
Burroughs
Clary
Computer Control Co.
Control Data Corporation
Digital Equipment Corp.
El-tronics, Inc.
'General Electric
General Precision
Honeywell Electronic Data
Processing
II-IV
46
Electronics, Inc.
NAME OF
COMPUTER
SOLID
STATE?
AVERAGE MONTHLY
RENTAL
EDP 900 system
Y
$7500
ASI 210
ASI 420
RECOMP II
RECadP III
205
220
EIOl-103
B250
B260
B270
B280
B5000
DE-60 IDE-60M
DDP-19
DDP-24
SPEC
G-15
G-20
160/160A
924/924A
1604/1604A
3600
6600
PDP-l
Y
Y
Y
Y
N
N
N
Y
Y
Y
PDP-4
y
PDP-5
y
ALWAC IIIE
210
215
225
235
LGP-21
LGP-30
L-3000
RPC-4000
N
H-290
H-400
H-800
H-1400
H-1800
DATAmatic 1000
HW-15K
2/61
y
y
$2850
$12.500
$2495
$1495
$4600
$14,000
$875
$4200
$3750
$7000
$6500
$16 200
$675
$2800
$2750
$800
$1000
$15,500
$1750/$3000
$11 ,000
$35,000
'$52,000
$120 ---DOD.
Sold only
about $120,000
Sold only
about $60,000
Sold only
about $25,000
$1820
$16,000
$5500
$7000
$10,900
$725
$1300
$45,000
$1875
y
Y
Y
y
y
$3000
$5000
$22,000
$14,000
$30,000 up
N
Y
$490
Y
Y
y
Y
Y
Y
N
y
Y
y
Y
Y
y
y
y
y
Y
y
y
semi
DATE OF FIRST
NUMBER OF
INSTALLATION INSTALLATIONS
4/62
2/63
11/58
6/61
1/54
10/58
1/56
11/61
11/62
7/62
7/62
4/63
2/60
6/61
-/63
5t:60
7/55
4161
5/60 & 7/61
8/61
1/60
6/63
NUMBER OF
UNFILLED
ORDERS
12"
10
8
1
125
32
70
55
Y60
54
33
16
20
5
105
3
'0
40
46
24
21
22
3
0
X
3
-3
10
345
22285
10
51
0
X
X
X
X
X
0
1
3
51
9
(}
11/60
42
5
1
9
8/62
13
14
11/63
0
2/54
7/59
-/63
1/61
-/64
12/62
9/56
1/60
1/61
32
74
0
120
0
17
395
1
68
6/60
12/61
12/60
5/64
-/63
12/57
6/63
10
53
52
0
0
5
1
2/64
D.
f'
X
6
17
80
3
33
5
0
17
3
64
7
2
2
X
2
COMPUTERS apd AUTOMATION for July, 1963
~
~
8
Ii II
§
~
I
I
I
~
I
§
il
...
WI
:II::
:i
CD
0
I-
:;:)
.c
~
.c
c
III
:;:)
D.
w
l)
~
no
...Z
0
I-
:I
8
0
te
C
:i:
M
fa
~
11\
00
.;
•
fa
:I
0
'D
~
";
C
0
iG»
Z
iZi
YES, start my subscription to COMPUTERS AND AUTOMATION according
to the instructions checked below, and
send me at once a reprint of "The Survey
and Study of the Computer Field."
If you are not at present a subscriber to COMPUTERS AND AUTOMATION, this is a good time
to come aboard, and join the growing numbers of
well-informed people who receive a personal copy
of C&A each month.
There is no risk! If at any time you feel that
COMPUTERS AND AUTOMATION is not supplying you with useful information and ideas worth
many times the cost of the magazine, tell us and
we will promptly refund to you the unmailed portion of your subscription. So may we hear from you?
~
:I
'"c-=
We welcome at all times the suggestions and
.:omments of friends, both new and old, on what we
do publish and what we should publish. We are a
small organization with the independence and flexibility to have an alert and patient ear.
•
z
0
Z
The excitement and progress that is everywhere in the computer and electronic data processing field is a continual incentive to all of us . . .
and we at C&A intend in the months ahead to
help stimulate some of the excitement and progress by presenting articles. ideas. concepts, and
graphics that reflect the vitality of the people who
are leading the "Computer Revolution."
As our "thank you" for responding now to this
invitation, we will send you at once a complete reprint of the "Survey and Study of the Computer
Field" by the Investment Bankers Association,
which appeared in the January and February issues.
A copy of this authoritative report will be sent to
you FREE if you act now and enter your subscription to the important issues of COMPUTERS AND
AUTOMATION coming up. A postage-paid reply
card is attached at right for your convenience.
~
I!I!
~I
GI
And if you are a regular reader of COMPUTERS
AND AUTOMATION you have probably noted the
recent improvements in our cover and typography.
~
; I
CHANGE . . . WHO US?
... no. not really ... We only improve!
1_=
o
I
One Year (including the June Computer Directory and Buyers' Guide: 12 issues) $15.00
D One Year (excluding the June issue: 1 1 issues)
$7.50
I
it
~I
I
Name:_ _ _ _ _ _ _ _ _ _ _ _ _ __
Title_'_ _ _ _ _ _ _ _ _ _ _ _ _ __
Organization_'_ _ _ _ _ _ _ _ _ _ __
Address_'_ _ _ _ _ _ _ _ _ _ _ _ __
City_'_ _ _ _ _ _ _ _ _ _ _ _ _ __
State_'_ _ _ _ _ _ _ _ _ _ _ _ _ __
EDMUND C. BERKELEY
Editor
0
0
0
0
Payment enclosed
Bill me
Renewal
New subscription
YES,
start my subscription to COMPUTERS AND AUTOMATION according
to the instructions checked below, and
send me at once a reprint of "The Survey
and Study of the Computer Field."
o
One Year (including the June Computer Directory and Buyers' Guide: 12 issues) $15.00
o
One Year (excluding the June issue: 11 issues)
$7.50
among the articles coming soon in
Name:._____________________________
COMPUTER MEMORIES:
WHERE ARE THEY HEADED?
Title:._ _ _ _ _ _ _ _ _ ._ _ _ _ __
Organization :. _______________________
THE ART OF THE PROCRAMMER
Address :_________________________
Ci~-·----------------------------
A PRIMER ON
PROGRAM DOCUMENTATION
State _.__________________________
o
o
o
o
Payment enclosed
COMPUTERS IN DEFENSE:
THE WHERES, WHYS, AND HOWS
Bill me
Renewal
New subscription
I
--------------------------~
I
I
I
t
I
I
J
I
I
I
co
VI
J
I
J
... as well as other interesting
and timely articles,
on the ideas
and by the people
that are leading the
development and use
of automatic information handling equipment.
~
Z
o
~
c:
-4
o
Subscribe now
~
and make sure
o
z
you don't miss
~
:j
these and other important
articles and reports in
5
."
m
0
VI
-I
0
~
~
~
VI
!II
m
~
~
=t
Z
:!!
~
VI
-I
P n
Vol
Vol
VI
Vol
~
VI
VI
I
a
I
II
in the coming months.
NAME OF
MANUFACTURER
NAME OF
COMPUTER
IBM
305
650-card
650-RAMAC
1401
1410
1440
1460
1620
701
7010
702
7030
704
7040
7044
705
7070, 2, 4
7080
709
7090
7094
7094 II
Information Systems, Inc.
ISI-609
ITT
7300 ADX
Monroe Calculating Machine Co. Monrobot IX
Monrobot XI
NCR - 102
- 304
- 310
- 315
- 390
National Cash Register Co.
Paekard Bell
P13
Pili 1co
Hudio Corp. of America
Scientific Data Systems Inc.
Thompson Ramo Wooldridge, Inc.
~IVAC
2~O
PB 440
1000
2000-212
-210, 211
Bizmac
RCA 301
RCA 501
RCA 601
SDS-910
SDS-920
TRW-230
RW-300
TRW-330
TRW-340
TRW-530
SOLID
STATE?
N
N
N
Y
Y
Y
Y
Y
N
Y
N
y
N
Y
y
N
Y
Y
N
y
y
y
y
y
N
Y
N
Y
Y
Y
AVERAGE MONTHLY
RENTAL
12/57
11/54
11/54
9/60
11/61
4/64
10/63
9/60
4/53
2/64
2/55
5/61
12/55
6/63
6/63
11/55
3/60
8/61
8/58
11/59
9/62
4/64
2/58
7/62
3/58
810
675
195
5950
12/60
$14,000
$2000
$0500
1/60
5/61
5/62
250
28
30
40
70
~/(>1
:ICJO
()/():!
0
0
$1Il~0
Y
Y
$1200
$:l:iOO
$ 70 to
Y
Y
N
$~2,OOO
$40,000
X -- no longer in production
COMPUTERS and AUTOMATION for July, 1963
I~~
0
0
1390
2
25
12
X
205
24
X
90
15
1
2
170
X
0
42
135
220
------'=1=-::5~
10
27
12
9
X
250
12
9
38
7
8
2
18
4
6
X
25
59
0
10
130
10
1475
Y
$8000
$26,000
$1500
$7200
8/58
12/61
2/63
9/63
550
6
100
0
N
Y
Y
$35,000
$45,000
$135,000
12/50
10/62
25
3
2
TOTALS
X
44
4
Y
Y
Y
~/60
X
X
262
94
6
16
9
0
37
10
0
17
53
12
12
65
865
$8500
$20,000
$15,000
$1200
Y
Y
3000
360
()50
22
250
X
Y
Y
N
N
Y
Y
X
X
32
Y
Y
Y
Y
Y
x
0
4
6
71
0
0
142
375
52
34
26
10
0
19
6
160
--- ---T~l6i)---------f~o
-/6:1
1/63
10/58
-/56
2/61
6/59
11/62
8/62
9/62
7/63
3/59
12/60
12/63
8/61
3/51 & 11/57
9/62
8/62
8/56
-/53
NUMBER OF
UNFILLED
ORDERS
4
$6000
$15,000
$35,000
$1700
$2690
$2680
$6000
$5000
$6000
$6000
Y
--~I~&~I~I=---------~N------~$~25~,000
Solid-State II
III
File Computers
60 & 120
Solid-state 80,
90, & Step
490
1004
1050
1100 Series (except 1107)
1107
LARC
NUMBER OF
INSTALLATIONS
$3600
$4000
$9000
$3500
$12,000
$1800
$9800
$2000
$5000
$19,175
$6900
$160,000
$32,000
$14,000
$26,000
$30,000
$24,000
$55,000
$40,000
$64,000
$70,000
$76,000
$4000
$35,000
Sold only $5800
$700
Y
Y
DATE OF FIRST
INSTALLATION
1G, 7 ()6
2
X
15
X
8252
47
CALENDAR OF COMING EVENTS
July 15-17, 1963: 3rd Annual Rochester Conference on
Data Acquisition and Processing in Medicine and Biology, Whipple Auditorium, Univ. of Rochester Medical
Center, Rochester, N. Y.; contact Kurt Enslein, 42 East
Ave., Rochester 4, N. Y.
July 22-26, 1963: 5th International Conference on Medical
Electronics, Liege, Belgium; contact Dr. L. E. Flory, RCA
Labs., Princeton, N. J.
Aug. 4-9, 1963: International Conference and Exhibit on
Aerospace Support, Sheraton-Park Hotel, Washington,
D. C.; contact F. K. Nichols, Air Defense Div. Directorate
of Operations, DSCjO Hdqs., USAF, Washington 25,
D.C.
Aug. 8-9, 1963: 6th Annual Summer Conference, Pacific
Science Center, Seattle, Wash.; contact Harold Ostling,
Secy., Northwest Computing Association, P. O. Box 836,
Seahurst, Wash.
Aug. 20-23, 1963: Western Elec. Show and Conference
(WESCON), Cow Palace, San Francisco, Calif.; contact
WESCON, 1435 La Cienega Blvd., Los Angeles, Calif.
Aug. 27-Sept. 4, 1963: 2nd International Congress on Automatic Control Swiss Industries Fair, Basle, Switzerland;
contact R. M. Emberson, Professional Groups Secretary,
IEEE, Box A, Lenox Hill Station, New York 21, N. Y.
Aug. 28-30, 1963: Associati~n for Computing Machinery,
Annual Meeting, Denver, Colo.
Sept. 9-11, 1963: 7th National Convention on Military
Electronics (MIL-E-CON 7), Shoreham Hotel, Washington, D. C.; contact L. D. Whitelock, Exhibits Chairman,
5614 Greentree Road, Bethesda 14, Md.
Sept. 9-12, 1963: 18th Annual ISA Instrument-Automation
Conference & Exhibit, McCormick Place, Chicago, Ill.
Sept. 9-12, 1963: International Symposium on Analog and
Digital Techniques Applied to Aeronautics, Liege, Belgium; contact M. Jean Florine, 50, Avenue F. D. Roosevelt, Brussels 5, Belgium.
Sept. 16-20, 1963: 2nd Institute on Electronic Information
Display Systems, The American University, SGPA, The
Center for Technology and Administration, 1901 F St.,
N.W., Washington 6, D:C.; contact Dr. Lowell H. Hattery, The American University, Washington 6, D. C.
Sept. 23-27, 1963: International Telemetering Conference,
London Hilton Hotel, London, England; contact F. G.
McGavock Associates, 3820 E. Colorado Blvd., Pasadena,
Calif.
Oct., 1963: 10th Annual Meeting, PGNS 2nd International
Symposium on Aerospace Nuclear Prop. and Power
Oct. 1-3, 1963: 8th Annual National Space Electronics Symposium, Hotel Fontainebleau, Miami Beach, Fla.; contact
Hugh E. Webber, Martin ,Co., Orlando, Fla.
Oct. 7-9, 1963: 9th National Communications Symposium,
Hotel Utica, Utica, N. Y.
Oct. 8-] I, 1963: Int'l on Electromagnetic Relays, Tohoku
University, Sendai, Japan; contact C. F. Cameron, School
of Eng., Oklahoma State University, Stillwater, Okla.
Oct. 14-15, 1963: Materials Handling Conference, Chamberlain Hotel, Newport News, Va.; contact R. C. Tench, •
C &'0 Rlwy Co., Rm. 803, ,C & 0 Bldg., Huntington 1,
W. Va.
Oct. 14-16, 1963: Systems and Procedures Association, 16th
International Systems Meeting, Hotel Schroeder, Milwaukee, Wis.; contact Systems & Procedures Association,
7890 Brookside Dr., Cleveland 38, Ohio
Oct. 17, 1963: 4th Annual Technical Symposium, University of Maryland, Baltimore, Md.; contact Hugh Nichols,
48
Dunlap and Associates, Inc., 7220 Wisconsin Ave., Bethesda, Md.
Oct. 21-23, 1963: East Coast Conference on Aerospace &
Navi,ga~ional Electronics (ECCANE), Balti!llore, Md.
Oct. 24-25, 1963: ~ymposium on Automatic Production in
Electrical and Electronic Engi~eering, The Institution of
Electrical Engineers, Savoy Place, London W. C. 2, England
Oct. 28-30, 196~: 19th Annual National Electronics Conference and Exhibition, McCormick Place, Chicago, Ill.;
contact Prof: Hansford W. Farris, Electrical Engineering
Dept., Univ. of Mich .. Ann Arbor, Mich.
·Oct. 28-Nov. 1, 1963: Business Equipment Manufacturers
Assn. Exposition and Conference, New York Coliseum, New
York, N. Y.; contact Richard L. Waddell, BEMA, 235 E.
42nd St., New York 17, N. Y.
Oct. 29-31, 1963: 10th Annual Mtg. PGNS 2nd Intn'l Symposium on Plasma Phenomena & Meas., El Cortez Hotel,
San Diego, Calif.; contact H. A. Thomas, Gen. Atomics,
Div., Geni. Dynamics, San Diego, Calif.
Nov. 4-6, 1963: NEREM (Northeast Research and Eng.
Meeting), Boston, Mass.; contact NEREM-IRE Boston
Office, 313 Washington St., Newton, Mass:
Nov. 10-15, 1963: 9th Annual Conference on Magnetism
and Magnetic Materials, Chalfonte-Haddon Hall, Atlantic City, N. J.; contact Mr. C. J. Kriessman, Physics, Materials and Processes Sec., Box 500, Blue Bell, Pa.
Nov. 12-14, 1963: Fall Joint Computer Conference, Las
Vegas Convention Center, Las Vegas, Nev.; contact Mr.
J. D. Madden, System Development. Corp., Santa Monica,
Calif.
Nov. 18-20, 1963: 1963 Radio Fall Meeting, Manger Hotel,
Rochester, N. Y.; contact EIA Engineering Dept., Room
2260, 11 W. 42 St., New York 36, N. Y.
Nov. 18-20, 1963: 16th Annual Conference on Engineering
in Medicine and 'Biology, Lord Baltimore Hotel, Baltimore, Md.; contact Richard Rimbach Associates, 933
Ridge Ave., 'Pittsburgh 12, Pa.
Nov. 19-21, 1963:' Fifth International Automation Congress
and Exposition, Sheraton Hotel, Philadelphia, Pa.; contact Internati('mal Automation Congress & Exposition,
Richard Rimbach Associates, Management, 933 Ridge
Ave., Pittsburgh 12, Pa.
Dec. 5-6, 1963: 14th Nat'l Conference on Vehicular Communications, Dallas, Tex.; contact A. C. Simmons, Comm.
Industries, Inc., 511 N. Akard, Dallas, Tex.
Feb. 3-7, 1964: ASTM International Conference on Materials, Sheraton Hotel, Philadelphia, Pa.; contact H. H.
Hamilton, American Society for Testinfjt and Materials,
1916 Race St., Philadelphia 3, Pa.
Fe,b. 5-7, 1964: 5th Winter Conv. on Military Electronics
(MILECON), Ambassador Hotel, Los Angeles, Calif.;
contact IEEE L. A. Office, 3600 Wilshire Blvd., Los Angeles, Calif.
Feb. 12-14, 1964: International Solid-States Circuits, Sheraton Hotel & Univ. of Pa.
Feb. 26-28, 1964: Scintillation and Semiconductor Counter
Symposium, Washington, D. C.
Mar. 23-26, 1964: IRE International Convention, Coliseum
and New York Hilton Hotel, New York, N. Y.; contact
E. K. Gannett, IRE Hdqs., 1 E. 79 St., New York 21, N. Y.
Apr. 22-24, 1964: SWIRECO (SW IRE Conf. and E1ec.
Show), Dallas Memorial Auditorium, Dallas, Tex.
COMPUTERS and AUTOMATION for July" 1963
BOOI(S AND
OTHER
PUBLICATIONS
attention of control engineers in an English-la-nguage text. In all, the text contains II chapters. Five appendices include: "Derivation of the Dual-input
Describing Function," and "An Analytic
Approach to the Inverse Describing Function." Index.
Moses M. Berlin
nurroughs Corp., StatE of the Technical
Training Dept" Radnor, Penna. I Digital Computer Principles I McGraw-Hili
Book Co., Inc., 330 W. 42nd St., New
York 36, N. Y. I 1962, printed, 5tl7 pp,
$10.50
.\ nonmathematical explanation of digital computers, their operations and the
components that make them work, is here
presented. The main objective of this
bc,)Qk is to evaluate the underlying concepts of computer logic and circuitry. It
has been written primarily for the computer engineer, programmer, and technician. The book is divided into three
sections as follows: "Computers and Transistors" (Chap. 1-7), "Computer Circuits"
(Chap. 8-15), and "Computer Units
(Chap. 16-24). Among the topics covered
in section I are: basic concepts of computer logic, symbolic notation, and number systems. Section II provides an analysis of basic computer circuits. Each circuit
is first analyzed with the vacuum tube as
the active clement and then the transistorized version of the circuit is analyzed.
Section III shows how the logical elements
described in the two previous sections are
connected to perform major computer
operations. Three appendices and i,ndex.
Allston, Mass.
We 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.
We shall be glad to report other information in future lists if a review
copy is sen t 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.
Nadler, Morton I Topics in Engineering
Logic I A Pergamon Press Book, The
Macmillan Co., 60 Fifth Ave., New York
11, N. Y. I 1962, printed, 231 pp, $9.50
This important book is based on lectures given at the Indian Statistical Institllle, Calcutta, in the Spring of 19:19,
which summarized the results of four
years' work at the Institute of Computing
Machinery in Prague. At Prague, the author came 'into contact with A. Svoboda's
methods for the design of synchronous
relay logical systems. Under this stimulus,
the author attempted to find electronic
systems with such properties that these
methods could be applied with little
change to the design of high-speed logical
systems. The result is this attempt at a
theory of synchronous electronic logical
systems in which close attention is paid
to the actual circuits employed. Seven
chapters include: "Physical Considerations: Elementary Logical Circuits for
Synchronous Systems," "The Minimal
Form," "Codes and Matrices," and "Structural Redundance." Chapter seven, "Examples," illustrates the application of various topics of the first six chapters in
actual design problems. List of references
and an index are included.
Gibson, John E. I Nonlinear Automatic
Control I McGraw-Hill Book Co., Inc.,
330 W. 42nd St., New York 36, N. Y. I
1963, printed, 585 pp, $16.50
The author of this work believes that
a general method of synthesis for nonlinear problems is impossible. However,
he believes one can develop a series of
analyses that apply to restricted classes
of nonlinearities and extend the usefulness of known techniques to a wider range
of systems. Such is the approach of this
text, written at the level of the first-year
graduate student. Chapter three, "Sampled-data Systems and the z Transform,"
consists of a discussion of a special integral transform called the z transform, and
modified z-transform techniques. It serves
as an introduction to the z-form method
disLllssed in Chapter 4, ":\lumerical :\lethods." Chapter 8, "The Second Method of
Liapllnov" is one of the first attempts to
bring' the Liapunov approach to the
Vancil, Richard F. I Leasing of Industrial
Equipment I :\[cGl'aw·HilI Book Co., Inc.,
3:1II W. 42 St., Nt'w York :Hi, N. Y. I 1!)(i:J,
printed, 2H:J pp, $12.511
This book }>n:sl'llls a ~pl'rilic tholOligh
examination of the C(,()1l01llits of c'llliplIICJlt
lcasing. It is writtell first, for thc l',('Cli'
tive responsible for making leasing dedsions. The first three chapters and the last
chapter give a thorough understanding of
the subject without the analytical procedure
to be used. The book is also addressed to
the analyst who must prepare the comparative evaluation of leasing plans. Chapters
four, five, and six give a detailed explana·
tion of the analytical procedure to be used.
Seven chapters include: "The Role of Leas·
ing in Capital Budgeting," "FUllltiollS Pel'
formed by Leasing Companies," "Financial
Leases: The Lease-or-BolTo\\' necision," and
"Economic Life and Residual Value." Con·
tains an appendix, "Intemal (Divisional) .-\(.
counting for Financial Leases," and a bibliography. .-\lso includes "Tables for rhe
Analysis of Financial .\lternatives and Capital Expenditures," pp 195-279-basically.
interest tables with some new tabulated
functions. Index.
H.
S. Gellman, editor, and 32 authors I
Proceedings of the Computing and Data
Processing Society of Canada I Univ. of
Toronto Press, Toronto 5, Ontario,
Canada I 1963, photooffset, 293 pp,
$6.00
This work contains the papers presented at the Third National Conference
of the Computing and Data Processing
Society of Canada, held at McGill University in :\[olltreal, June 11-12, 1962. The
theme of the conference was "The Computer as an Aid to :\1anagetnent." Thirty
papers are included in this work. among
which are: "Decision-Making Using a
Computer: A Transportation Company,"
"Central Control of One Million Parts
Locations," "Heat Exchanger Design,"
and "Computer Evolution to Aid Compilers."
We're looking for engineers who get
restless resting on their laurels, who
are anxious to move on to the next
achievement, who are as excited as
we are about the future of airborne
digital computers,
In 1959 we produced the LC-600, an
airborne digital computer with acapability comparable to that of a large
ground-based machine. Its computational power: 10,000 additions per second. We were not content. We looked
for more performance, less size and
weight. We got it. Our LC-820 airborne
digital computer is capable of 250,000
additions per second. Weight: 124
pounds. Volume: 2.3 cubic feet.
Our objectives for the future are to
decrease size, weight, and cost even
further while increasing reliability.
Will you be on the team that packs
more performance in a smaller, more
compact computer? You will if you
feel as we do about the future of
aerospace computers.
If you're ready for a step ahead in
the airborne digital computer field
and/or inertial systems, look into litton. Simply send your name and
address for an application form or your
resume for immediate action. Write to
Mr. J. B. Lacy, Guidance and Control
Systems Division, 5500 Canoga Avenue, Woodland Hills, California. An
equal opportunity employer.
[8
LITTON SYSTEMS, INC.
Guidance and Control Systems Division
COMPUTERS and AUTOMATION for July, 1963
tl,9
NEW PATENTS
RA YMOND R. SKOLNICK
Reg. Patent Agent
Ford Inst. Co., Div. of Sperry Rand
Corp., Long Island City 1, New York
The following is a compilation
of patents pertaining to computer
and associated equipment from the
"Official Gazette of the U. S. Patent Office," dates of issue as indicated. Each entry consists of patent
number / inventor(s) / assignee /
invention. Printed copies of patents may be obtained from the U. S.·
Commissioner of Patents, ''''ashington 25, D. C., at a cost of 25 cents
each.
April 2, 1963
3,083,903 / Russell H. Larson, Wappingers
Falls, ~. Y. / I.B.M. Corp., New York,
N. Y., a corp. of N. Y. / Data Trans.J!!ting System.
DATA PROCESSING EQUIPMENT
EXCHANGE COMPANY OFFERS:
705 system, $100,000.
650 System, 8 machines, 4 years
old, $100,000.00.
Maintenance Approximately $950.OO/Month. 4-727's and 355 RAMAC
could be added' on.
604-521, $6500.00; 031 KP $350.00; 2-082's; 2- LGP 30's; 085 Collator.
WANTED
Used Analog Computers; 1401's;
KP's, Sorters, Collators, Tabs, SP's.
Send facts about any good used
DP machines.
BUY or SELL through
DA-PEX Company
366 Francis Building
Louisville, Kentucky
GL 1-7457
JU 5-5454
:1,OH3,910 / George ~1. Berkin / Pough-
. keepsie, N. Y. / LB.~r. Corp., New York,
~. Y.• a corp. of N. Y. / Serial Adder
alld Subtracter.
3.084.286 / Robert E. Leo. SUlln)'slope.
.\riz. / General Electric Co .• a corp. of
~ew York / Binary Counter.
3.084.33:1 / William J. Greene. Scotch
Plains, ~. 1- / .-\ir Reduction Co., InL,
~ew York, N. Y .. a corp. of N. Y. /
~Iethod and .-\-pparatus for Transmitting In telligence.
3.084.334 / Louis H. ~Iartin. Concord. and
Edward 1- Lucas, Cochituate, ~Iass. /
.-\ vco Corp .• Cincinnati. Ohio. a corp. of
Delaware / Direct Access Photomemory
for Storage and Retrieval of Information.
3.084.336 / Donald G. Clemons. Newark.
N.1- / Bell Telephone Labs .• Inc.. New
York. N. Y.• a corp. of N. Y. / Magnetic
Memory Construction and Circuits.
April 9, 1963
3.084.854 / Henry F. Schunk. Wakefield.
and Joseph 1\1. Welty. Natick, Mass. /
Automatic Records, Inc.. Natick, Mass .•
a corp. of Mass. / Data Processing System.
3.08·1,859 / Otto J- M. Smith. C01~tra Costa
County. Calif. (612. Euclid Ave .. Berkeley. Calif.) / - - - / Number Storage
Apparatus and Method.
3.084.860 / Ben B. Jordan. Watchung,
N. 1- / Western Electric Co .• Inc.. New
York. N. Y .• a corp. of N. Y. / Decimal
to Binary Number Translating Device.
3.084.861 / Allen W. Roberts. South Plainfield. N. 1- / Bell Telephone Labs .• Inc..
a coPp. of New York / Logic Circuitry.
April 16, 1963
3.086.118 / Frederick Arthur Summerlin.
Isleworth. England / The Sperry Gyroscope Co .• Ltd., Brentford. Middlesex.
England. a company of Great Britain /
In tegra ting Devices.
3.086.197 / ' John L. Anderson. Poughkeepsie. N. Y. / LB.M. Corp .• New York.
N. Y .• a corp. of N. Y. / Cyrogenic
~remory System.
April 23, 1963
Theodor E. Einsele. Sin delfingen, Germany / LB.M. Corp .• New
York. N. Y., a corp. of N. Y. / Data
Processing Machine.
3,086.708 / Howard Berkowitz. Cook
County, Ill .• and Bernard W. Moss,
Baltimore County, Md. / The MartinMarietta Corp .• Baltimore, Md .. a corp.
of Maryland / Method and Apparatus
for Automatic Digital Process Control.
3,087.074 / William N. Carroll. Rhinebeck, Roderick A. Coopper, Hyde l)ark,
3.086,706 /
I
and Richard C. Counihan. Kingston •
N. Y. / LB.M. Corp .. New York. N. Y .•
a corp. of N. Y. / Transistorized Logic
Circuit Operative in the Pulse ~Iode.
3,OH7.076 / Carl O. Pillgry. III. Lexington •
Ky. / Lfi.M. Corp., ~ew York. N. Y .• a
corp. of N. Y. / Logic and/or Gate
Having ~Iagneticall)' Induced Pulse~ as
One Input.
3.087.142 / Robert D. Buck, Saugerties,
~. Y. / J.fi.M. Corp., New York. N. Y .•
a corp. of ,N. Y. / Binary Data Transfer
Apparatus.
3.087.149 / Jack W. Malcolm. Gettysburg.
Pa. / The National Cash Register Co.,
Dayton, Ohio. a corp. of Maryland /
Decimal to Binary Conversion and Storage System.
April 30, 1963
3.088.056 / Ronald Tevonian. Flemington.
N. 1- / Western Electric Co .• Inc., New
York. N. Y., a corp. of N. Y. / Logic
and NIemory Circuit Units.
3,088,099 / Wilbur E. Du Vall, Gardena.
Calif. / The W. W. Henry Co., Huntington Park, Calif.. a corp. of Calif. /
Data Communication System.
3.088.100 / Joseph W. Crownover, La
Jolla, Calif. / Litton Systems. Inc.. Beverly Hills. Calif.. a corp. of Maryland /
Diodcless Magnetic Shift Register.
3,088.103 / Leon Gryk. New Britain.
Conn. / Royal McBee Corp., Port Chester. N. Y., a corp. of New York / Matrix
EncodC'fs.
May 7, 1963
3.088,688 / Abraham Harel. Framingham.
~Iass. / Radio Corp. of America, a corp.
of Delaware / Binary Adder Employing
Minority Logic.
3,088.671 / Robert L. Chase, Blue Point,
~. Y. / U.S.A. as representated by the
U.S. Atomic Energy Commission / Multiplier Circuit.
3.089,124 / Glenn _E. Hagen. Manhattan
Beach. CharleS" R. Williams, Palo
Verdes. Arthur Y. Baker, Hermosa
Beach, and Robert E. Jackson. Redondo Beach, Calif. / Alwac International Inc.. a corp. of Panama / Computer System with High Capacity Random Access Memory.
3.089.126 / James C. Miller, Hamilton
Square. N. J. / Radio Corp. of America.
a corp. of Delaware / Negative Resistance Diode Memory.
3.089.127 / Leon J. ~rintz, Brooklyn. N. Y .•
and Roland Yii, West Chester. Pa. /
Burroughs Corp .. Detroit, Mich., a corp.
of Mich. / Magnetic Shift Register.
3,089,128 / Charles B. Smith, Vestal. N. Y.
/ LB.M. Corp .• New York, N. Y .• a
corp. of N. Y. / Magnetic Core Switching Circuit.
ADVERTISING INDEX
Following is the index of advertisements. Each item contains: Name and address of the advertiser / page number
where the advertisement appears / name of agency if any.
Addressograph-Multigraph Corp., 1200 Babbitt Rd., Cleveland 17, Ohio / Page 4 / The Griswold-Eshleman Co.
American Telephone & Telegraph Co., 195 Broadway, New
York 7, N. Y. / Page 2 / N. W. Ayer & Son, Inc.
Bellcomm, Inc., 1100 17th St., N. W., Washington 6, D. C. /
Page 51 / N. W. Ayer & Son, Inc.
Control Data Corp., 8100 34th Ave., So., Minneapolis 20,
l\linn. / Pages 24, 25 / Erwin Wasey, Ruthrauff & Ryan, Inc.
Data Processing Equipment Exchange Co., 366 Francis
Bldg., Louisville 2, Ky. / Page 50/ -.
International Business Machines Corp., 590 Madison Ave.,
New York 22, N. Y. /Page 31/ Benton & Bowles, Inc.
International Business Machines Corp., Data Systems Div.,
50
Box 390, Poughkeepsie, N. Y. /Page 21/ Benton & Bowles, :nco
Litton Industries, Guidance & Control Systems Div., 5500 Canoga Ave., Woodland Hills, Calif. / Page 49/ Ellington & Co,
National Cash Register Co., Main & K Sts., Dayton 9, Ohio /
Page 19/ McCann-Erickson, Inc.
Photocircuits Corporation, Glen Cove, N. Y. / Page 6 / DuncanBrooks, Inc.
Presto Seal Manufacturing Corp., 37-27 33rd St., Long Islanq
City, N. Y. / Page 15 / Technical Operations, Inc .• 3600 M St., N. W., Washington,
D. C. / Page 30/ Edwin F. Hall
Valley Consultants, Inc., 716 York Rd., Towson 4, Md. / Page
30 / George C. Ruehl, Jr.
COMPUTERS and AUTOMATION for July, 1963
\
.1
(CoQ
Acurrent Bellcomm project
The Moon, Mars, Venus, long objects of wonder, are now destined for exploration ...... As you read this, Bellcomm is at work for NASA-planning
early phases of this vast exploration, analyzing systems needed for landing
man on the moon ...... If you're qualified, you might like to join the vanguard
of the expedition. There are rewarding openings in such fields as computing
and programming, physics, mathematics, engineering, man-machine relationships, aerodynamics and aeronautical engineering ...... Bellcomm, the newest
company of the Bell System, is an equal opportunity employer located in
Washington, D. C. Resumes will be promptly and carefully considered. Address
them to Mr. W. W. Braunwarth, Personnel Director, Bellcomm, Inc., Room
1101U, 1100 17th Street, N.W., Washington 6, D. C......
4
BELLCOMM, INC.
A Bell Telephone System Company
here it is!
W~rID9g) W~lID ~[ffi ~~® ~(Q)~~mM~®W ff~®~~C9[ffi~/~~~
A new comprehensive edition of the WHO'S WHO IN THE COMPUTER FIELD is available. This is the first edition in
over five years. All
... entries are complete and accurate as of January, 1963. This handsome, clothbound book is
the standard biographical reference on over 5000 leaders In computer applications/ design/ education/ logic/
mathematics/ marketing/ programming/ systems analysis
'J
This is the answer book for such questions as:
Where did he get his degree?
How do you spell his name?
What is his home address?
Where is he working?
What is his job title?
What are his interests in the computer field?
What papers has he given recently?
What books has he written or edited?
To what societies does he belong?
L
Order your copy now of this valuable volume. The supply is definitely limited.
To order your copy send a check or purchase order for $24.95 to:
Who's Who in the Computer Field, Attn: Order Section,
815 Washington Street, Newtonville 60, Mass.
~ffl?M~j£if.i£~i~1.7f.::'
¥i~i~;:~J~;:Fi!?::r:L~f::;
iSi,i!i,j:=ff;[:ff,ilr'i!i;fif!! .
E::'H7:Jf:!0~ffi:!!3f.~}f~;;
))4
- - - - - - - - - -_ _ _ _ _ _..r:r.::~::.E
Source Exif Data:
File Type : PDF File Type Extension : pdf MIME Type : application/pdf PDF Version : 1.3 Linearized : No XMP Toolkit : Adobe XMP Core 4.2.1-c043 52.372728, 2009/01/18-15:56:37 Producer : Adobe Acrobat 9.1 Paper Capture Plug-in Modify Date : 2009:03:19 20:59:15-07:00 Create Date : 2009:03:19 20:59:15-07:00 Metadata Date : 2009:03:19 20:59:15-07:00 Format : application/pdf Document ID : uuid:e6ea9a9b-7719-45a0-a603-2d892cde22a2 Instance ID : uuid:a7314213-0789-43b2-b00f-6829c5af3707 Page Layout : SinglePage Page Mode : UseNone Page Count : 56EXIF Metadata provided by EXIF.tools