195602

195602 195602

User Manual: 195602

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

Download
Open PDF In Browser	View PDF

The Function of Automatic Programming for Computers
in Business Data Processing
R. E. Rossheim
Computers and Engineering Education

Paul E. Stanley

The Pldnning Behind the IBM 702 Installation at
Chrysler Corporation
. . . Eugene Lindstrom
Publications for Business on Automatic Computers:
A Supplemental Listing
Ned Chapin
Automatic Mixup

. . . lawrence M. Clark

UNIVAC: "You're trying to divide by zero"
A scientist, testing a formula on Univac®
recently, was amazed to see the computing system stop, then automatically
type the reproof: "You're trying to divide by zero." A quick check proved
that Univac, as always, was right.
This graphic demonstration points
out just one of many Remington Rand
refinements in the art of computer programming and operation. For Univac
has been trained to spot human errors.
It can now carry out, commands given
in simple business English. It can even
manufacture automatically its own pro-

gram of instructions - at electronic
speeds, with unequalled accuracy.
These skills have been developed
through Univac's unique experience in
electroriic data-processing. Because,
with every Univac delivered goes 10
years' experience in electronic computing ... 5 years' experience in the commercial type of data-processing. This
wealth of background in programming
and operation is unobtainable elsewhere.
The unprecedented savings of Univac
data-processing have been proved by
solving actual customer problems-not

DIVISION OF SPERRY RAND

- 2· -

CORPORATION

by working out theoretical solutions
with non-existent computers. When you'
install Univac, you're sure to get under
way faster, surer, and more economically because the System has already
handled similar work.
Univac is now at work in leading
organizations throughout the country.
And, in today's competitive market, the
company which cuts its overhead first
comes out on top. So don't wait until
1957 ... 1958 ... or 1959 to cash in on
the tremendous savings available to you
now with the Univac System.

COMPUTERS

ROBOTS

CYBERNETICS
Vo 1. 5, No.

AND

AUTOMATION

•

AUT,OMATIC

CONTROL

2

February,

1956

ESTABLISHED SEPTEMBER. 1951

ARTICLES AND PAPERS
••• R. J. Rossheim

The Function of Automatic Programmdng for
Computers in Business Data Processing
Computers and Engineering Education
The Planning Behind the IBM 702 Installation
at Chrysler
Publications for Business on Automatic
Computers: A Supplemental Listing

6

••• P. E. Stanley
••• E. Lindstrom

10
13

••• Ned Chapin

16

••• R. R. Skolnick

32

REFERENCE INFORMATION
New Patents
Association for Computing Machinery Meeting: Names and
Addresses of Authors of Papers

34

FICTION
Automatic Mixup

••• Lawrence M. Clark

21

••• B. Danch

40

FORU\t

Another Programmdng Failure

4
4

The Editor's Notes
Index of Notices
Advertising Index

54

Editor:
Edmund C. Berkeley
Assistant ~ditors:
Eva ni~tefano, Jewell Bown,
Neil D. Macdonald, F.L Walker.
C:ontribu ting Edi tors: Andrew D. Booth, John

).I.

Advisory Committee:
Samuel B. Williams.
Herbert F. Mitchell. Jr •• Justin
Oppenheim

Breen, John W. Carr. III. Alston S. Householder, Fletcher Pratt

Publisher: Berkeley Enterprises. Inc.
Publication Office:
513 Avenue of the Americas, New York 21, N.Y. - Algonquin 5-7177
Research and Library:
36 West 11 Street •• New York, 11. N.Y. - Gramercy 7-1157
Branch Office: R15 Washington Street., Newtonville 69, Mass. - Decatur 2-5453 or 2-3928
Advertising Representatives: San Francisco - W. A. Babcock, 605 Market St., San Francisco 5, Calif.
Los Angeles - Wentworth F. Green. 439 So. Western Ave., Los Angeles 5, Calif.
elsewhere - the Publisher
o)MPlJfERS AND AUTOMATION is pub 1 ished monthl y. Copyright. 1955 by Berkel ey Enterp ri ses. Inc. Subscrip tion rates:
$5.50 for one year, $10.50 for two years, in the United-States;
$6.00 for one year,
$11.50 for two years,
in
Cana"da;
$6.50 for one year, $12. ~ for two years elsewhere. Bulk subscription rates:
see page 46. Advertising ra tes:
see page 44-.
Entere(i as second class matter at the Post Office. New York. N. Y.

- 3 -

THE

THE

C~1PUTER

EDITOR'S

DIRECTORY

The June 1956 issue of "Computers and AutomatioQ" l'1ill be the second issue of "The Comprter Directory". It will have three parts: Partl,
"Roster of Organizations in the Computer Field";
Part 2, "The Computing Machinery Field: Products
and Services for Sale"; Part 3, "Who's Who in
the Computer Field". For more information alnut
entries, etc., see the notice in this issue on
the Computer Directory.

ADDRESSES OF GIVERS OF PAPERS
At the Association for Computing Machinery
meeting in Philadelphia, in September 1955, ~
120 papers l1ere given. In the November issue cf
"Computers and Automation", we printed the tites
and abstracts, using the meeting program mrl~
ing by photooffset. Since the program did not
shol\' the addresses of the givers of papers, I'Ve
received a number of requests from readers of
"Computers and Automation" asking hol'{ they could
~rlte to the speakers.
We found out that the
addresses had been dittoed by Professor John P.
Nash, chairman of the Program Commdttee of the
Association for Computing Machinery, and he has
kindly given us this list for printing. In th~
issue, therefore, we print the names and ~jdre~
es of persons giving papers at that September
meeting.
We hope that future programs of meetings
will ShOlf enough of the address of each person
giving a pap~r, so that persons interested in
papers can write to, the givers of papers, in~e
i'lterim before publication·, and obtain more information about the paper than is given in the
abstract.

NOTES

wrong address, send us your correct address,
marking it AP so that we know this address is
to match up with the mailing list used for "adv~nce programs" of meetings.
This mailing list of computer people is
being maintained by "Computers and Automation"
tdth the help of the Joint Computer Conference.
This list is not released for advertising purposes: it is used (1) by us from time to time
to improve our "Who's Who in the Computer Fleld~'
and (2) by the Joint Computer Conference from
time to time to send out notices of their meetings. Other similar organiZations may of .couae
use it, at ~mall costs, for mailing out meeting
notices and advance program mailings.

CORRECTIONS
In the September issue of "Computers and
Automation", on page 17, after the name of the
author (I. Asimov~ and before the start of the
story ('tFranchise"), there should have appeared the following note: (A longer version of
this story appeared in the August 1955 issue
of IF magazine; copyright, 1955, by Quinn Publishing Co., Inc.)

In the October issue of "Computers and
Automation", on page 30, in the "Who's Who"
the name Burton Grad, Production S peci ali st.,
was incorrectly listed as Burton Brad. We regret this unfortunate error.

INDEX- OF NOTICES
For Information on:

MAILING LIST OF COMPUTER PEOPLE
For the mailing of the Western Joint Computer Conference for the meeting in San Francisco, February 1955, I'Ve provided about 11,500
names of computer people, produced from IB M
punch cards. The slips Ifere made by an IBM 40;
Tabulator and enclosed in a window envelope.
There are undoubtedly errors on these cards: if
the slIp'you received contained an error in yar
name ana address, we should greatly appreciate
your sending us the erroneous slip, with the
correction marked. If that slip is no longer
available, but you remember that it showed a
- 4 -

Advertising Index
Advertising Rates and
Specifications
Back Copies
Bulk Subscription Rates
Computer Directory
Corrections
Manuscripts
Reader's Inquiry Form
Special Issues

See Page:
46
44
42

38
37
4

36
46
38

Address Changes: If your address changes,
please notify us giving both old and new
addresses, and allOtoJ three loJeeks for the change.

NOW!
dependable relays for printed circuits
Maybe you, too, have been awaiting availability of a
good relay for direct insertion into printed circuits. Now
Automatic Electric can solve your problem with a miniature relay that is just right.
120 million operations, without a single readjustment
or relubrication! That's what you get from. this rugged,
improved Series SQD Relay, because it features a special
heavy-duty bearing and bearing pin. Also a recess in the
bearing plate retains an adequate supply of lubricant
for long-term lubrication of the bearing pin.'
Consider these additional advantages:
1. The sections of the terminals that insert into the
printed circuit board are NOT brazed or welded
into place, but are integral parts of the coil terminals and contact springs-thus preventing IDternal loss in conductivity or continuity.
2. Terminal design permits direct plug-in of the relay
into a printed circuit board, ready to be secured-in
place with any acceptable soldering technique.
Usually the desired contact spring combination, or
pile-up, is sufficiently large so that additional mounting
(support) of the relay is not necessary.

-5-

SQD Miniature Printed Circuit Relays are available
with many different contact spring arrangements, and
for a multitude of applications. Springs can be made of
phosphor-bronze, ((Bronco" metal, or other specialpurpose materials, as required.
Of course the long life, heavy-duty features of the
improved SQD Relay can be had in the conventional
type of plug-in relay, if regular sockets are preferred for
use, whether in printed circuitry or other applications.
To get complete details, write: Automatic Electric Sales
Corporation, 1033 West Van Buren St., Chicago ,7,
Illinois. In Canada: Automatic Electric (Canada) Lt~.,
Toronto. Offices in principal cities.

RELAYS

'"""" "
AIJTDMATI[

',"'"' '

SWITCHES

"'"""'" , "
ELE[TRI[

CHICAGO

The Function of Automatic Programming' for Computers
in Business Data Processing
R. J. ROSSHEIM
Supervisor, Industrial Pro~rammin~ Research
Rand, Division of Sperry Rand r~rp.,
~iladelphia 3, Pa.,

Remi~ton

(Presented before the Association for Colll>uting Machinery, Philadelphia, Pennsylvania, Sept. 14

This paper is concerned td th the role that
automatic programming systems for large, general-purpose computers might play in the application of these equipments to business d a t aprocessing problems. My definition of automatic programming for the purpose of this discussion is extremely broad, including any programming system which is capable of substantially
reducing the time and effort now required to
program the machines. The definition of automatic programming certainly includes the co ncept of relegation of purely clerical portions
of machine coding to the computer itself, but
it is not limited to this concept.
It is hoped that this p~per will stimula~
interest and encourage effort in developing rrore
advanced automatic programming systems for business use.
Training
I shall s tart with the training ph a s e
tVhich normally precedes the computer installation. I shall make the assumption, t.ghich is
now virtually a cl iche, that the company embarlting on the computer venture has decided to train
its own people in the use of the equipment, rather than to educate trained programmers from
outside the company in the endless detail of
its data-processing procedures. The immediate
problem is to train people in the use of a machine which far transcends in terms of f Ie x ibility, power, and speed anything to which t~
might be accustomed. Frequently, we req«ire
that the people selected have superior ability
in overall system planning since we expect the
equipment to have a sizable impact on our procedures, and yet the firs t thing these p eo pIe
must do is submerge themselves in the swirling
mul ti tudes of electronic binaries which m u s t
eventually be arranged, bit by bit, to control
the execution of the payroll accounting func~
or the like. The basic problem is that the
equipment when delivered has buil t into it only
one basic ability, that being the ability to
interpret a certain language which is the i ns truc tion code of the machine. This code c e rtainly is logically complete and can be adapted to meet every conceivable requirement of the
user. At the same time, it is elementary, or
stated another way, it is a language of monosyllables. An indicator of the size of these
elements is the revelation in more than ore computer installation that it requires many tens
of thousands of these elementary machine i n-

- 6 -

',0

16,

NSS)

s truc tions to· describe completely the' payroll
accounting function.
The training problem would be considerably
reduced if the equipment were able to interpre t
the everyday language of the procedure analyst
and to translate the meaning so conveyed into
its own binary arrangements. This is, of course,
an ideal. It is fundamentally dependent on the
existence of such a language, assuming also
precise meanings and universality. Theconcept
of such a language is not so new, nor so ideal,
that there are not even now some res ear c h
groups t1l'0rking in this area. I bel ieve the
objective can be stated as raising the level
of machine comprehension to the language of the
user, rather than forcing him to adop t mmosyllabic machine talk.
The abili ty to program would b ecom e a
relatively simple operational tool similar to
data flow charting, or procedure manual writing.
It would permi t the procedure ana lys t or his
counterpart to spend more of his time taking
the broad view of data flow through a system,
and to concentrate on the objective of good
system design.
I think that very few of the people selected to work in this field will be i nte rested in becoming professional programmers;
so the requirement is to make the equipmen t
available to such people on a basis which is
cons is tent with the other tools 0 f the i r
trade.
I am suggesting that the obj e c t i v e of
automatic programming techniques duri n g the
initial training stages might be the development of a relatively simple pseudo-code \\fhich
would employ to tihatever extent now possible,
the language of the user. The computer would
be enabled to translate programs wri t ten in
this language into the elementary mac h i n e
codes. The irnroodiate effec t should be t hat
trainees could learn in much less time towri te
usable programs. At the same time the t endency would be away from creating highly-trained
specialis ts and toward the development of parttime or temporary programmers from those who
are experts in the jobs wh ich are to be prepared for the machines.
I should note that, although there is not
to my knowledge a pseudo-code system for business applications, an experiment involving the
same principle tvas successfully carried out by

FUNCTION OF AUTOMATIC
o

the Whirlwind computer group at M.LT. two
years ago. The so-called Summer Session Computer t'las a programmed modification of Whirlwind t'lhich permi tted the principles 0 f P r 0gramming to be taught to a student g r 0 up of
varied interes ts and backgrounds, based 0 n a
pseudo-code which was easily learned and applied.

P~OG~AMMING

Curren tly, it is too cos tly to try a number of
different approaches to the same application,
if this involves programming a second tim e.
Moreover, the systems we design are likely to
be very conservative in order to red u c e the
chance of having to make maj or changes. Thus,
tve miss many opportuni ties to learn more about
how to use this new equipment as the pressure
of schedules forces us to settle upon procedures which may be far from optimum.

Approximation and Experimentation
Next, I will consider the role of au t 0matic programming in the time consuming and
costly preparation of data-processing procedures for a computer system. Before discussing
automatic programming per se in this reg ard,
I will talk about programming procedures fo r
commercial livork.
First, the concept of approximation. Th~
approach is suggested by the tremendous amount
of detailed information t'lhich must be gathered,
analyzed, organized, and finally progr a mmed
into a computer data-processing procedure.
Information may come from procedure man u als,
rules, and regulations which are sometimes
subj ec t to interpretation, or from infor med,
but sometimes inarticulate clerical personnel.
As a result of the nature of the informatio n
and its sources, a fundamental requirement of
programming cannot be met; that is, the "problem", or job, cannot be completely predefined.
Therefore, programming must often begin tV it h
incomplete or incorrect information, and the
firs t program is necessarily merely a fir s t
approximation. Frequently, voices which were
silent during the survey and fact-finding visits, are quick to provide the necessary criticism tvhen the first approximation misses the
mark. Then the hard reali ties begin to be revealed, the neat flow charts spread out, d eveloping all manner of intricate networks, and
the programs s tart to overflow a va i 1 a b I e
storage. If I have conveyed the idea at all,
it should not be necess ary to say t hat the
second approximation is not final, nor the
third; but finally an acceptable, if not perfect, version is developed which comprehends
everything of maj or importance as well as most
minor details. The fact is, although the approximation technique described above may not
be consciously employed, it represents a good
description of what actually happens.
Another approach, which is the other face
of the same coin, involves the idea of experimentation. Whereas approximation met hod s
convey the idea of working gradually toward a
single correct and best procedure, the experimentation viewpoint acknmdedges the f a c t
that our analytical tools for designing electronic data-processing systems are extreme 1 y
primi tive, so that our current p rogres s i s
largely the result of trial and error meth~.

If these ideas are reasonable, involving
as they do the use of approximation tech~es
and experimentation in the design of commercial
data-processing programs for computers, automatic programming can play several import ant
roles. In fact, it is apparent that any systematic approach to programming which w 0 u 1 d
assist in the process of going from on e approximation to the next without recoding would
reduce the total amount of coding considerably.
Flow Charting
I shall now discuss the various ph a s e s
through which an application or job must pass
before a working program is ready for the computer. It is significant, I think, that problems which have arisen in all of the are a s
could be solved, either wholly or in part, by
one or more of the techniques tvhich falltdthin
my definition of automatic programming.
Firs t, let us consider flow c h art in g •
Apparently the flow chart, in a mul ti tude 0 f
forms, is the most easily adapted and universally accepted method of describing d a t aprocessing. The need is for some way 0 f arranging great masses of detail in such a way
that we can see both forest and trees at the
same time. Typically we proceed fro m the
broadest level of flow charting tvWeh outlines
the entire application, dOlivu through sever al
successive levels, each adding detail, unti 1
in some cases, we dratv charts which virtua lly
indicate the machine instructions required to
perform each block on the chart. Such charts,
I have heard it said, could be handed to people
tvho have been trained in the computer instruction code, and they could translate the detailed flow chart into a machine pro g ram
wi thout us ing any discretion or j u d g men t.
They tvould perform a clerical function. But
why not let the computer itself perform the
clerical function of translating detailed flm"
charts into sequences of machine instructions?
I believe that the translation of flow charts
into machine codes should be a primary objective of an automatic programming system for
business data-processing.
Some of the problems in trying to do this,
are very challenging. First, we need a "flowchart-to-digi tal-computer" converter so t hat

- 7 -

FUNCTION OF AUTOMATIC PtlOGRAMMING

we can feed the flow chart into the computer.
Second, any auto-programming sys tern I can conceive of must work from a very s pee i f i e
"language" that must have all the precision of
the machine code, while at the s arne time adapting itself to the requirements of flow charts,
of programmers, and of the applications. I will
not stop to enumerate other problems fu accomplishing what I have suggested, but perhaps I
should make one point clear. It is certainly
true that many programmers leave the flow chart
behind in the pressure to write machine codes,
but remember that what I am suggesting should
replace the wri ting of machine codes. I a cknowledge, too, the fact that often when th e
programmer and clerical coder are the sam e
person, as is more often the case than not, the
detailed flow chart is omitted entirely. However, I would expec t that as we learn m 0 r e
about the elements of data-processing, we will
be able to build auto-programming s y s tern s
capable of interpreting higher 1 eve 1 flo w
charts, Ivherein each block represents a functional subroutine containing many machine instructions, rather than just one or two.
My
feeling is that the first step in this direction
is the really difficult one.
Coding
In the conventional sequence, after flow
charting comes coding, and since flow chartmg
is not going to replace coding tomorrow, there
is considerable l"lork for auto-programming in
the coding phase. It is in this area wher e
currently some encouraging progress is be in g
made. I shall merely mention some of the contributions that auto-programming systems can
make to business data-processing applications.

First, automatic programming systems can
assist in standardizing the programs by specifying the form of the coding, and by supplying a library of subroutines which are copied
into programs Ivherever they app ly • Sec 0 n d,
as more is learned about recurrent subroutines,
libraries of such routines can be develo p ed,
and programs can be written in a shorthan d
pseudo-code, thereby avoiding duplication 0 f
coding which has been Ivritten before, andconsequently reducing the amount of netv coding in
each nel"1 program. Third, the tedious boo kkeeping jobs associated with the preparation
of finished coding can be wholly 0 r par t 1 y
relegated to the computer, thereby releas i n g
the programmer to concentrate on the de s i g n
of a complete and efficient computer procedure,
rather than burdening him with coding detai~.
The fact that not enough is notv k n 0 tV n
about the technical language of data-processing should not discourage the development 0 f
initial automatic coding systems. As certain
processing functions are defined, standard rou-

- 6 -

tines can be written and pseudo-codes invented
to call up these routines. However, now andfor
some time to come, any automatic ..coding system
must accept conventional machine coding as well
as pseudo-coding.

Another improvement in programming p r 0cedures which can be designed into an aummatic
sys tern is the abili ty to have several p e opl e
lyork on different parts of the same program
without the usual problems of communication.
Program Checking or Debugging
After the program has been prepared, it
must be checked or debugged, which frequently
requires many additional man-hours and ~hine­
hours. Much of the abusive language concerning the cost of application of computer s i s
directed at this stage of the operati 0 n.
Basically, two kinds of prograrmning err 0 r s
account for the effort which is expended a t
this point. First, there are the simpl eminded errors resulting from poor or inc 0 mplete record-keeping during the pro c e s s of
coding, or from slight misunderstandings of the
way in which the equipment operates, 0 r th e
typographical kind of error. Second, there
are the logical errors which may resul t fro m
incorrect translation of the flow chart, 0 r
failure of the programmer to consider certain
cases or combinations of cases. It is v er y
easy to sympathize with the programmer for he
must have the ability to work in minute de~il
and, at the same time, keep in mind the co mplex interrelationships wi thin the e n t i r e
framework of the program. Putting s ym path y
aside, h01vever, there is a pressing need to
improve methods of prograrfiming in or d e r to
avoid some of the errors, and at the same time
to provide for efficient methods of err 0 rcorrection. This means that the aut om a ti c
programming system does not stop functio nin g
wi th the 'production of the first "approximate"
program. Systematic error correction tecmdques
should be provided to assist correction of all
kinds and levels of errors. Some study should
be made of debugging procedures con sid e ring
typical business installations, the operating
characteristics of the equipment, and the requirements of the programmer.
The availability of an automatic programming system may avoid many of the pit fall s
which arise in the debugging stage. For instance, if the approximation idea~ utilized,
it is conceivable that a programmer would prepare a routine concerned only with the m a i n
flow of data, knowing at the time that t her e
l'Vere many loose ends. By debugging the p a rtial routine he has the advantage of working
with a smaller, less complex program initially.
Once this is checked out, it should be re 1 atively simple to control and isolate errors

FUNCTION DF AllTOMATIC PROGRA~IMING
as additional routines are added to make the
the program to fit into the existing organizaprocedure complete. Here too, automatic protion, without the complete recoding which congramming techniques l'1ould substantially reduce
ventional methods would require.
the time and effort, since it should be possible
to blend in new subroutines and correct existRelated to the foregoing is the ide a 0 f
ing routines, taking bes t advantage of the parts
revision of the program, at some later d ate,
of the program which are complete, cor r e c t,
to provide for changes in the procedures. I t
and tested.
is a mistake, I think, to regard business applications as static. Just as business itself
With respect to the testing phase, the
is dynamic, the data-processing requireme n t s
most important contribution an auto-programmmg
are constantly changing, and it would see m
system could make is the reduction in overa 1 1
reasonable to rate a data-processing system in
elapsed time required for the first approximaterms of its ability to respond to c han g e d
tion. It is vitally important to reach at the
conditions. Automatic coding could be the key
earlies t possible date the firs t check poi n t
to this ability.
l..nere the critical look is taken and the information feedback is activated. No degree of
This migh t be an appropriate poi n t to
accuracy and no amount of care can replace fue
mention another problem. When a programmer is
first test in turning up weaknesses in the proin the midst of a particular routine he can
gram or procedure.
follow the coding h~ has written quite easily,
but let him move on to another coding as s ig nment, and after a week he will have I a r gel y
Testing
forgotten the first program. It is almost a s
difficul t for him to trace through the cod in g
Once the program has been debugged to the
as for someone who has not seen i~ before. Any
point where it is operable, it must be checked
programming system should provide a method for
against the other contingent parts of the syscarrying along verbal descriptions of the functions which the different parts of a pr ogr am
tem. Actual operating conditions and d a t a
must at least be simulated. How heartbreaking
perform. If this could be tied in to the flow
are the programming failures at this poi n t !
chart, subroutine by subroutine, together they
~Yet, this often is the firs t real checkpo i n t
lfiQuld provide adequate documentation 0 f the
in the sense that some ac tual data are proprogram. The verbal description would reve a 1,
cessed' and the programmed procedure is m 0 r e
to non-programmers exactly what the pr 0 gr a m
easily checked to see whether it fulfills all
does, and it could be used for checking p u rthe requirements between input and 0 u t put.
poses. Moreover, it would greatly facilitate
This is the first approximation. Since it is
revision of the program weeks or months later.
very inconvenient to check coding itself, itis
necessary to process input and to produce outBefore closing, it might be well to menput, both of which must be referred t o t he
tion one other area which may soon draw the
people in the business who either origi na t e
attention of management, if it has not already
source documents or use the reports. The s e
done so. Much is currently being written and
are the important critics of the procedures.
said about the ability of electronic data-proThey are the inspectors at the end of the processors to provide information to assist ma' nduction line. Since the procedure was originagement decisions. One form this might tak e
ally set down and the initial flow charts were
is the processing of files and records already
drawn there has been considerable dead-reckonstored in a high-speed medium. There are two
ing. Using conventional methods of programmbg
important aspects of this type of job. First,
and coding, there has also been considera b 1 e
time is of the essence; and I am speaking 0 f
time. Now the feedback of information begins,
the time from the original request to the time
and the procedure begins to be refined. The
the results are turned over to manage me n t
second, third, and successive approximatio n s
Second, it cannot be assumed that these requests
are written, each requiring repetition of some
could be anticipated in the sense that programs
or all of the preceding steps.
could be prepared ahead of the request. At the
same time, programming costs cannot be exorbitant as such requests may be for one-shot jobs.
Clearly, this is a challenge which could bernet
Corrections, Modifications, Revisions
by an adequate automatic programming system.
After the results of the first approximation have been checked a steady flow of modifications, corrections, and additional routines
Conclusion
must be merged in with the original program.
All too often the delicate balance is upset by
In the short space of this paper I h a v e
these changes, and succeeding debugging is even
tried to show why the techniques of automatic
more difficult than the original. Hopef u lly,
programming are important throughout business
an automatic system would permit new parts 0 f
data-processing application of large-s c a Ie,
(cmtinued on page 32)

- 9 -

COMPUT!ER~S,

AND

ENGINEERING

EDUCATION

PAUL E. STANLEY
Associate Professor of Aeronautical Engineering,
Purdue tniversity
Lafayette, Indiana

Anyone who has attempted to make a ch~
in an engineering curriculum is n'ell at'ltare of
the multitude of problems. So many new ideas,
. -inethods-, -and machines have come to the fore,
each clamoring for a place in the curriculum,
that one is tempted to eliminate them all and
return to the teaching of only "bas ic physical
principles" and "fundamental engineering design", t'lthatever that may mean. HOt1Tever, this
ostrich in the sand attitude is not conducive
to progress; one must somehow eliminate some
of the traditional subject matter and introduce the new. (The readrr is referred to the
"Saber Tooth Curriculum" for some other thoujrts
along this line.)

for emphasizing the similarity 0 f s y s t ems,
whether they are electrical, mechanical, 0 r
thermal, etc. The integration of many fundamental principles thus can be achieved ina
fashion long sought by engineering instructors.
The same can be said for many of the mathematical principles taught by the mathematics department but so frequently neglected in the
teaching of other courses.
Every engineering student knows well (we
hope) the relation F rna, force equals mas s
times acceleration. If he has had a course in
vibrations, he may be aware that

=

,

~

Ma..:::MU

cit.%.
and be able to Ivri te, tv! th some understanding,
the differential equation

It is not the purpose of this paper to
say what must be Id thdral'1tn from the curricu1um but rather to suggest a nen' subject. An
attempt will be made to shOt'lt how the net" idea
can be used in conjunction tdth the traditional.

NJ.2Ix 1- fdx +

cit"

The large scale advent of automatic computing machinery into the field of engin eering
research and development 11.rarrants careful examination by engineering educators.
In the
aeronautical industry, and to
great extent
~n other areas, both analog and digi tal computers often confront the young engineer within the first few months of his new career. It
seems reasonable, therefore, for him at least
to have been introduced to this important unl
during his undergraduate days.

a

The research and development groups i n
industry use both digital and analog computers.
Perhaps it Ivould be good to try to inc 1 u d e
something of each in engineering instruction.
This naturally calls for both types of moch~
to be available for use by the undergraduate.
Often this is not feasible for, even if computers of advanced types are located on the
campus, they are employed exclusively in graduate instruction and research. This is especially true of digital computers; but, as lvill
be shOt'ltn, analog equipment suitabl.e for instruction can be built (or bought) for a very
reasonable sum.
In addition to providing some familiarization Id th computing machinery, the use 0 f
the analog computer provides an excellent ~

ott

Kx ~ F(t)

But fet" students, if any, are aware 0 f the
similari ty betlveen this and the ele c tr i c a I
equation
'

L J,t
tL~ + RL

T

..LSLd-t
= E(t)
C

The less familiar form of the latter equation
is obviously

L ol,:tct, -+ R~ + L
J,tl.

cit

q.::.

c'

E. (t)

which is identical mathematically to the mechanical equation.
Such lack of. understanding, of which this
is but one example, definitely restricts the
creative work tvhich an engineer might oti1elWise
be able to do. The solution of a few prob~,
similar to those above, on the analog compurer
serves well to emphasize the. mathematical similarity and brings to light again the usefulness of simple calculus and differential equation principles.
A not-so-incidental product of instructiDn
in the use of the analog computer is the use
of the notation and principles of operational
calculus. In more advanced courses, the concept of the transfer function can be intxoduced
and the analysis (and synthesis) of s ys tern s
such as simple servomechanisms can be carried
out. A system can then be simulated 0 n the

- 10 -

COMPUTERS AND ENGINEERING EDUCATION

computer and an expe imental anaiysis of its
behavior determined.
Examples of this will
be given later.

.

2

The implication of the above is this: by
use of the basic elements of the analog computer, a model of a system, any sys tem uooer study, can be constructed and analyzed. Since tre
sys terns s tud~ed i.n undergraduate classes are
usually simple~ the number of analo~ elements
is small, consisting of a few DC amp lifiers
and power supplies; some capacitors for integration; resistors for in put and summing ~es;
a few potentiometers and a recording galvanometer for the output. By proper selection of
these elements, and flexible arrangements 0 f
them, valuable ins truc tion in many areas can
be given.
The stability of the equipment and the (£curacy of the results are, to a great extent,
dependent upon the quality of the components
used. Leakage resistance in the integrating
capaci tors, or from various components to grourrl
must be kept as high as possible, highrom~d
to the input and feedback resistors which are
us ually of the order of .1 to 1 megohm. fbvever, a little care in construction of the
equipment and some selec tion of parts fro m
commercial grade stock will provide us able
equipment at minimum cost.

- Q() = Lor; Ol
0(

.

+ M~a + Mi 9 + M~~

tl1here m is the mass of the aircraft; v is the
airspeed; e is the angle of pi tch; 0( the is an amplifier lvi th capaci tive feedback, thus
is an integrator (with 3 inputs);

Figure 2

is an amplifier with resistance feedback used
as an inverter only.

It will have the open loop transfer
A series of runs determining 0{, , for
different Jed and edi ted these r eports and prepared a methods and procedures mar
ual. The manual lias completed in August 1953,
and has proved to be a most effective way to
educate persons working on the project. Besides.
such a study is valuable in itself since it re-examines procedures within an organization. Because
procedures have a tendency to remain s ta tic
while the l~ork to be done continually changes,
revieh' from time to time is essential to keep
operations as profitable as they might be.
The Parts Division 1'11as particularly investigated and studied, in the light of the pr~
dures outlined in the manual, and certain a pplications 1>Jere selected which would apparently
lead to an integrated system of operation thnugh
the use of electronic data processing m~hinery.
In Oc tober 1953 poss ible programs lW'ere outlined
and graphically illustrated with flow charts.
Finally, preliminary phase 1vas conclud~d b y
Chrysler's action ordering an IBM Type 702.

In order to make necessary changes in policy, reorganize departments, and convert procedures, a committee called the Executive Policy Committee l'11aS appointed. This committee
consisted of all of the managers of the various departments affected by the project; its
first step was to examine all of the proposed
changes recommended by the planning g r 0 up.
Where policy and procedure changes suggested
by the planning committee were not approved ~
the executive committee, a compromise between
the two groups was reached.
The changes in policies and procedures
were made official by the issuance of written
procedures approved by the Division Comptrol~r
and the Vice President of the Parts Division.
By mutual agreement the application to central
inventory control was selected in July 1954 as
the first task to be assumed by electronic wta
processing machinery (EDPM).
CENTRAL INVENTORY CONTROL
In central inventory control, before mechanization, a certain "stock status record"
made with bookkeeping machines reflected the
activity which occurred from eight to twelve
weeks prior to the posting •. The processi n 9
cycle for this bookkeeping machine method was
monthly. In order to prepare for EDPM, the
operation was converted to a punched card system. The punched card operation summarized
activity weekly and produced an Availability
Report which indicated the status of inventory
as of b.ro to three weeks prior. Us ing the 7(Q,
the processing cycle of the present system has
become daily: it reflects the -inventory activity which has occurred during the preceding dey.

- 13 -

IBM 702 INSTALLATION
item should continue to be stocked in a particular area where sales have declined. If th e
prerequisite conditions for stocking materials
are not satisfied, the machine prints 0 uta
stock s t"atus report which tells management why
the report was printed and what action is to
be taken. A third type of review made by the
machine is for follow-up ac tion. If the current availability satisfies national and local
requirements but the "on Hand" inventory is leg;
than 60 or 30 days' supply for both requirements, then a stock status report is printed
out and directed to the expediter for followup action.

This permi ts "management by exception," that is,
isolation of those critical areas that nee d
special and immediate personal attention outof
a maze of 250,000 routine decisions each day.
TIlree basic tapes serve as input for the
inventory control analysis. The first is the
sales record tape which is a by-product of the
central invoicing operation. The second tape
contains all of the transactions except sales.
The third is the master stock status Ie d g e r
tape. The purpose of this particular run ism
produce two reports and the updated master inventory ledger tape; these ttVO reports will control the procurement and distribution of the inventory. The first is a stock status report,
comprehensive and produced on an exception basis. Whenever this report is issued some type
of action is required. It contains all of the
information necessary to complete any action
tii thout reference to any other record. The second report or distribution report is used to
allocate material due in at the central fortvarding station for distribution to other warehouses, and to release back orders. Both r eports are printed on a tape-to-printer operation t\'1 th the output tape records generated by
the 702 in this operation. The mas ter s to c k
status ledger tape is automatically updated as
a resul t of this operation. That is, all indicative data that has changed and inventor y
status of each part are automatically recorded
on the new master stock status ledger tape.
Once the machine has completely posted the
stock status record for one part number the next
job is to analyze the stock status report and
decide if any action has to be taken on t his
item. In th is program an average of t tV e n t yfour decisions must be made by the 702 for each
part that has had activity. The average day's
business includes ten thousand active par t s.
This" means that 240,000 decisions -- busin es s
decisions, not program steps -- are made during
the regular tt\'o-hour run each day.
The decisions fall into four basic categories:
0)
procurement; (2) distribution;
expediting; (4) disposition of surplus.

(3)

These are the four broad classifications of the
type of decisions that the machine makes on eam
part every time it is active. The 702 comp~te­
ly watches all condi tions and also checks the
levels of stock and the demand being placed on
outlying plants for parts. There are a number
of different formulas for acquisition of parts
depending on the distribution policy for t h 'e
type of part, service ruling, or cost. The IBM
702 examines much of the heading data in the record and decides tvhat parts of this data apply
to the particular part being considered. The
machine may even determine Ivhether or not a n

The distribution report, the second re~
of this particular run, is also used for the
allocation of parts received which are on bac~
order. The 702 automatically allocates the
quantity to be used against back orders by date
or age until the entire quantity is exhausted.
Should it be necessary to make a partial shipment because the allocated quantity is not sufficient to cover all back orders, the 702 will
automatically record this case. This particular record is maintained on the back-order summary tape. The quantity of parts on back order
may vary from several hundred to a thousand. It
is expected that this number will be greatly
reduced through the daily control of inventory.
To aid the back-order situation, a weekly customer shortage report of all back orders i s
also prepared. TIle historical data now being
recorded on magnetic tape should be quite valuable in the future use of the 702 for comprehensive analyses of inventory and stock movements. TIlis will enable management to exe rcise more precise methods in proj ec ting future
inventory movement.
CENTRAL INVOICING
TIle second application is central invoicJng. To make this operation easier each tvare:house has an installation of electric accounting (punch card) machines I-vhich prepare- shipping orders. TIlis enables shipments tore made
ordinarily within eight hours after receipt of
a customer's order. Notification of shipment
is received from the shipping department, the
cards which were used to prepare the shipping
order are pulled, changes are made in the cards
,to indicate the exact quantity shipped, and
correct shipment cards including name and address cards and data cards are forwarded daily
by air express to the central invoicing cen ter~
A control card is inserted with each group of
cards indicating the exact amount of sale for
that group. When the cards are received they
are automatically balanced to the control card
and converted to magnetic tape.

- 14 -

The following output tapes are prepared
as the result of the invoicing operation:

IB~

702 INSTALLATION

The invoice record, converted by a tapeto-printer operation, preparing the invoices.
1.

Hi th no action being required in the way of key
punching. However, four basic classifications
have been established for handling remittances
automatically or semi-automatically:

2. The invoice register tape, converted
by the printer, preparing the invoice register.

1. Complete cash payment with the complete discount honored.

The sales analysis tape by product, by
type of customer.
3.

2.

4.

The accounts receivable tape.

5. Disbursements tape, discussed in the
inventory control application.
6.

Partial cash remittance with the dis-

count.
3.

Complete payment with no discount.

4.

Partial payment with no discount.

Back-order detail for releasing back

orders.

The cards are then sorted to dealer n urn be r,
merged tvi th miscellaneous transaction and adj ustment cards, and converted to magnetic tape.
This tape becomes the input remi ttance tap e •
Each day the accounts receivable are made current by up-dating the mas ter tape. As a r esuIt it is possible for the Credit Department
to have a daily record of all accounts exceeding their limits.

The invoicing operation is currently being
performed w~th the use of electric accounting
machines installed at five warehouses. When
this operation is c.onverted to the 702 and performed centrally it will reduce the t,"ork load
in these five installations by 5~/o. Invoicing
for the entire United States is done in two
hours of 702 time with an additional 12 hours
for the tape-to-printer operation each day.

~

COS T OF SALES

The third application is cost of sales.
Unit cost for each part is available on the
master inventory ledger tape. Disbursements
are extended to write a tape for cost of sales
f or the day's sales; then they are merged with
the month-to-date cost-of-sales tape. At the
end of the month the final cost of sales is
printed on a tape-to-printer operation so that
immediately after the close of the month's business the cost of sales for all sales for that
period is available. This is an excellent example of the extensive by-product advantages
that result from the integration of an operation of this type.

AlES ANALYS IS

The fifth application is sales analysis.
The various distributions and summarizations
required for sales analysis are easily obtained on the ,702 by reading the tapes into the
central processing unit. Based upon this information a large number of distributions can
be stored on the magnetic drum. A quarterly
report is prepared by dealer and accoun ti n g
class showing the amount of money, the number
of orders by type of order, and whether it was
a stock or emergency order. Accurate. __sales
statistics for control of dealer sales are always available becaus@ of this operation. An
annual report is prepared by warehouse and by
states for all delivered parts.
CONCLtEIONS

ACCOUNTS RECEIVABLE
The fourth application is accounts receivable. It is interesting to note that the ~
or portion of Chrysler Accounts Receivable results from the sale of parts. These tiansactions are recorded on magnet1c tape produc e d
during the invoicing operation. The balan c e
is key punched into cards and then convert ed
to magnetic tape. The 702 is used for too sorting of the tape into dealer sequence for t w <>
groups kn01'ln as the 85% group and the 15% group,
and the two tapes are merged prior to the accounts receivable run. The remi ttance tap e
used as input is written at the time a statement is sent to a dealer or a customer. A pr~
punched remi ttance card is enclosed 1\'1 th the
request that the card be returned with the remittance. Thus most of the cards are returned

This integrated data processing s y s ten,
has become a valuable tool for the management
of Chrysler. The immediate advantages thatthe
company may expect from the use of electronic
data processing machines are these:
1. The major portion of all input information is available as the result of a previous
automatic operation.

2. Management are able to use their time
more economically, because the logical ability
of the electronic data processor permits a discriminating selection of more critical probrems
for study.
3.
tained.

Operation at high speed has been ob(continued on page 32)

- 15 -

PUBLICA TIONS

FOR

BUSINESS

ON

AUTOMATIC

COMPUTERS:

A SUPPLEMENTAL LISTING
NED CHAPIN

Lecturer, Illinois Insti tute of Technology
Q.icago, illinois

This paper lists for business people w hat
migh t be called a supplemental reading 1 i s t 0 f
publications in the field of automatic computers.
This listing is supplemental to the one published
in "Computers and Automation" for Sept. 1955 with
the title "Publications for Business on Automatic
Computers: A Basic Listing." It is expected that
this supplemental listing will be followed by a
third paper t'li th the title "Reference Listing."
The publications listed in this paper present
additional points of view and additional information (as t\'e11 as repeating much of the same information) on the same topics covered by the pub 1 ications ci ted in the "Basic Listing." The publications cited in this supplemental lis ting are
grouped by subject matter. Within each subject
mat ter grouping, the publications are 1 is te din
approximate order of scope and depth of coverage,
the more elementary being listed first, the mor e
advanced being listed last. Items that are background material have been identified as such in the
annotation.

Use of Computers in General
Walter II. Wheeler, "Wanted: A Revolution in the
Office," Management Review, V 42, #12 (Dec. 1953),
pp. 712-713
Tedious pape~~ork should be mechanized.
LmITrence P. Lessing, "Computers in Business,"
Scientific American, V 190, #1 (Jan. 1954), pp 2125
An easy-to-read general introductory article.
Everett S. Calhoun, "The Challenge of Electronic
Equipment," 'NACA Bulletin, V 34, Section I, # 10
(June 1953), pp. 1267-1279
A general introductory article.
Matt W. Boz, "You, Electronics, and the Brass,"
Papers of the Sixth Annual Systems Meeting (N e h'
York: Systems and Procedures Association of America, 1953), pp. 14-20
Some good suggestions from one systems man to
another on investigating possible automati c
computer uses.
P. D. Laubach and L. E. Thompson, "Electronic Computers: A Progress Report) "Harvard Bus iness Review,
V 33, #2 (March-April, 1955), pp. 120-128
Cost savings are slow to appear from the use
of automatic computers, this survey indicates.
The artic Ie also re-emphas izes that to use
an automatic computer requires a good de a 1
of preliminary work.
C. E. Knight and C. II. Fawkner, "The Impact of
Automation on the Company Organization," Ge ner a 1
Management Series Number 178 (New York: Amer ican
Management ASSOCIation, Inc., 1955), pp. 11-21
A computer Hill have an important r ole i n
- 16 -

automated plants, and organiZational changes
will have to be made to reflect that role.
William Bamert, "Making A Feasibility Stu d y:
Strengthened Management Control," Electronic Data
Processing in Industry (New York: American Management Association, Inc., 1955), pp. 112-124
This article lists some of the advantages ,in
terms of systems improvements that might b e
realized with the use of an automatic computer.
Paul Kircher, "The Gap Between the Ele c t ron i c s
Engineer and the Accountant," The Controller, V22,
#8 (Aug. 1954), pp. 358-362, 374
.
Vocabulary and atti tude are the causes of the
gap; better communication between th e two
groups should be established.
James Gibbons, "Is Push Button Accounting Aro un d
the Corner?" Interal Audi tor, V 11, #1 (M arc h ,
1954), pp. 44-53
A payroll example is given to help the author
anSlver "not yet."
John H. Lindholm et al., Electronic Business Machines -- A New Tool for Management {Winches te r,
Mass.: John H. Lindholm, 1953),47 pp.
Also, in
condensed form: "What Business Can Expect of Electronic Office Machines," The Office, V 38, #6 (~c.
1953), pp. 81, 85-86, 89-90, 92
Automatic computers are not giant brains, but
they can nevertheless still be used in business.
F. J. Porter, "EEl, Con Edison, and Electronics,"
Edison Electric Institute Bulletin, V 22, #12 (Dec.
1954), pp. 419-424
Describes the preparation these groups h a v e
mal.le for the use of automatic computers.
Joseph E. Perry, et al., "ABA Report of the Co mmittee on Electronics," Computers and Automation,
V 3, #1 (Jan. 1954) pp. 10-12
The report sugges t a rather conservative gradualism in adopting electronic equipme n t in
banking.
Feasibility Studies
Radio Corporation of America, An Electronic A ccounting System for a Medium Size Utility (N e w
York: Radio Corp. of America, June 1953), 86 p~.
This is a summary of a feasibility study.
Ned Chapin, "Justifying the Use of An Automa tic
Computer," Computers and Automation, V 4, #8 (Aug.
1955), pp. 17-19
Expected savings must be balanced against increased inves tment in justifying an automatic
computer.
Matt W. Boz, "Building a Checklis t for Dec i s ion
on Electronic Accounting," Office Management Series Number 131 (New York: American Management Association, Inc., 1953), pp. 9-13
Discusses some of the considerations important to an oil company in the possible appli-

PllULICATIONS
cation of automatic computers.
handling System," Electronic Data Processing in
John F. Feagler, et a1., "Making a Feas i bi 1 i t y
Industry (New York: American Management AssociaStudy, " Elec tronic Data Process ing in In d us try
tion, Inc., 1955), pp. 84-96
(New York: American Management Association, Inc.,
A minimum of information should be t ran s1955), pp. 74-124
mitted, but it must be adequate in content,
A discussion of the Chesapeake and Ohio Railform, accuracy, and timing.
way Co. 's approach to the uses of an autormtic
Kenneth A: Wulff, "Planning for the Automatic Ofcomputer.
fice--Consul tant 's Vielvpoint," Workshop for Ma nR. F. Clippinger, "Economics of the Digi tal Co magement (New York: Management Magazines, Inc. ,
puter, " Harvard Business Review, V 33, 11: 1 (Jan.1955), pp. 349-354
Feb. 1955), pp. 77-88
Attention should be given to systems changes.
Discusses in general terms the cost of operWarren P. Livingston, "Where Do We Stand Today on
ating an automatic computer.
Automatic Accounting Machines," Office Management
Emerson F. Cooley, "Elec tronic Equipment Appl i e d
and Equipment, V 15, 11:1 (Jan. 1954), pp. 24 -26,
to Periodic Billing," Computers and Automation,
83-84
V 2, 11: 8 (Nov. 1953), pp. 17-21
Not much of anywhere until systems and methIf a systems analysis can show that a co s t
ods are standardized.
saving will result, then the use of an autoRalph II. Eidem, "Review of Developments in Office
Electronics," Office Management Series Number 135
matic computer may be desirable.
Bernard Whitney, "How to Sell Accounting Systems,"
(New York: American Management Association, Inc.,
1953), pp. 3-9
Systems and Procedures Quarterly, V4, #4 (Nov.
1953), pp. 6-7, 22
Even though present equipment is neither ideal
Do not try to do it with the "accountant's
nor perfect, it can be used to advantage.
cost data.
II. D. Buskey and V. R. Huskey, "Nelli Frontiers in
Joseph 0. Harrison, "Use of Computing Machines in
Business Management and Control Being Established
Operation Research," Operations Research for Manby Electronic Computers," Journal of Accountancy,
V 93, 11:1 (Jan. 1952), pp. 69-75
agement vay of
remembering it; three, a way of operating tdth
it; and fourth, a way of getting that information out of the machine, fast enough and reliably enough so as to direct shipments and traffic. All of these processes have now had over
40 years of development. The reliability and
speed in these devices is now uptvards of 100,000
times that of a human being.
"For example, tve now have Equipment t hat
uses pulses spaced one hundredth of ami c r 0second apart -- that is, a hundredth of a millionth of a second. These have been standard
operating pulses for computing equipment fo r
the last 20 years. Here is the speed Ive tV'ant
and need for automatic traffic control.
"Let me give you one more example. tv e
have almost unlimited information storage capacity, magnetic books, sheets of magnetic surfaces on which information is stored as an arrangement of magnetized spots. So information
can be stored at the rate of a billion decimru
digits in a cubic foot ,of space. We can read
any desired item of information in about ahu~
dredth of a second. This also has been demonstrated to be reliable over th~ last 20 years.
"Mr. CrulIllV'ell, there is no doub.t Ivhatever
of the capacity of science to build and operate automatic traffic control equipment which
is fast and reliable. It will be much faster
and more reliable than human beings. All that
t'll'e have to do is to put together es tablis h e d
and proved devices, to solve a problem that is
becoming more and more important. But t~ must
not forget that some errors, some pieces of
trouble, are inevitable, no matter how excellent the equipment.
CrulIllvell said in his loud voice, "Yeah,

tvell, you say what they all say. Eberhard, does
this man know tvhat he is talking about?"
Eberhard said, "I believe he does."
Crumwell said: "How many of the rest of
you think so?" Six of the other eight men sai:l,
"Yes, I do."

~vhat

Crumwell said, "Sam, you did not say yes-do you think?"

Eberh ard tvh is pered to Duns ter: "Th at's
Sam Oliver, head of the 2nd International Ban~
Netv Chicago office."
Sam Oliver said, "What I know about llinstEr
is hearsay, t'll'hich hOlvever agrees in reporting
Dunster to be an expert. But I know Eberhard,
and if Eberhard is tdlling to give adequate security, I think we can go ahead with the first
installment. After all, if we begin with New
Chicago and environs, and make a success of th at,
then we can spread out."
CrulIllvell got up: "OK, Eberhard, let's begin ld th your Natrasys and I tvill loan t t1 0 thirds of the commitment for the first year am
then we tvill see."
Eberhard got up, smiling and rubbing his
hands: "Thank you, Mr. Crurmvell. We're all
going to do tvell." Crurm.vell nodded goodbye and
went out in haste.
The meeting broke up. Dunster said t
Eberhard: "How tvell do you know CrulIll'lell?"

0

"Andy, not too well, but well enough. When
he thinks he can make a dollar, and nothing has
yet gone tvrong, he is OK. Once we get started,
we'll run on our own steam."
"Harry, I do hope you're right. I don't
like him. Besides I don't think he even heard
my remark about errors."
"Nobody really likes him, Andy, but he's
useful. tt
The first year had gone unexpectedly well,
but the team of Eberhard and Duns ter did 'n 0 t
finish tv! th needing Mr. Valerian CrulIl'tTell. The
mora Natrasys grel'\1' , the larger became CrullMell's
investment, less in proportion but greater in
amount. And as usual he required ample securi ty, in this case an extensive mortgage on Eoorhard's Sagamore Railway. It took all the spellbinding that Eberhard could muster over his
stockholders to keep that going.
For those
over, Dunster,
-- tvhen he was
fair amount of
- 24 -

first six years, that were now
with Eberhard's staunch backing
not distracted
and wi th a
Crumwell's money, had tv 0 rk e d
--I

long hours lvi th a fine crew of roon.

AUTOMATIC MIXUP

trouble by installing lead shielding in the
Radiation Laboratory and in the Natrasys Building.

They had built the principles of automatic
control of traffic into intelligent machinery.
From the first, the automatic traffic management service was a success, and it had spread
rapidly in a wider and wider network 0 u t 0 f
Nel'V Chicago. On the campus of Agamek University, next to MacNeil's Computation Laboratory,
the great Central Control office of Na t r a sy s
occupied the second floor of the Natrasys BWUding. Panels of the equipment overflowed into
the third and fourth floors, ld th thick c 0 nnee ting cables. The machine - if i nd e e d it
could be called "a" machine -- was quite probably one of the most intricate and intelligent
assemblies of automatic control machinery yet
in existence.
.

As a regular routine monthly test, Dunster
had arranged that a Natrasys test crew would
hunt for the nelvest radiation apparatus in any
of the nearby Radiation and Physics Laboratory
buildings once a month. They would lift upthe
shielding around the apparatus, direct a beam
of radiation at the Natrasys building, and test
to see if any pulses got through Nat r as y s
shielding. The last thirteen monthly testshad
been pulse-proof.
As a general measure to forestall trouble,
Dunster had considered duplicating the eqWpmem
in Natrasys, and requiring that both of two twin
pieces of equipment agree on the same information before it was released to some other part
of the system to use. This was still too expensive for Natrasys so far.

There had been troubles from time to time.
Usually, Valerian Crumvell found out about the
troub~~ too, and-then he raised-~ storm, each
storm ~orst than the last one.

But Dunster had installed another protection throughout Natrasys. This was that Natra-·
sys would carry along a check on any piece of
information equal to the sum of the digits and
letters (letters having the values A 1, B 2 up
to Z 26). For example, piece of informa tion
Y2254 would have associated with it as a check,
the number 38, namely the sum of 25, 2, 2, 5,
and 4. Any change in a single character would
alter the sum, although a pair of compensating
errors, such as an interchange of two digits,
would not be detected by this protection. This
had worked remarkably well, and enabled automatic elimination of a large number of errors.

Once there was a connecting wire that was
slightly loose, and swinging. Every now and
then it l'lould stdng against a terminal t hat
contained pulses 1'lith the value 6. Fin ally,
the insulation on that connecting wi r e wore
through at that point, and so every now and then
a 6 liould corne into some calculation where i t
had no business at all. If it had not been for
a caravan of ar~ trucks that provided a cons iderable amount of ground-shaking for a hal f
hour one morning, Dunster wondered how they ever
would have found out about that trouble.
Another time, tvhen the system was s t i l l
small and the disruption l\'as minor, there had
been extensive random misbehavior, completely
unexplained_Jor about a day. Then Dunste r
correlated it with the aurora borealis and an
unusually pOlverful soiar magnetic storm.
He
had prevented similar trouble in the.future by
appropriate shielding of sensitive parts of the
equipment.

III.
TIle telephone rang, and Dunster abruptly
returned to the present. He picked up the receiver.

Another trouble occurred when a new section of Natrasys equipment t'V'as buil t in Indianapolis. This trouble however tl/as a "s af e "
one -- it never permi t ted the equipment to work
right even once until it was removed. It turned
out tpat one of the young ladies lfielding a
sold~ting iron had disliked the red-yellolv-red
color-coded resistors because of her anti-Spa~
ish feelings, and had put in p a tr i 0 tic redlihi te-blue resistors.
Once, the new Radiation Laboratory to the
southh'est had been trying out some netv radiation apparatus. Extra pulses were 1 ib er all y
sprinkled all through the Central Cont r 0 1 Office. Dunster guessed this one almost at once
because he had been thinking something like this
might readily occur. lIe had liquidated thi s

"Professor Dunster", said the g u a r d's
voice, "Mr. El>erhard has arrived, and is coming
at once to your office."
As Dunster put the receiver down, Eberhrurl
opened the door to the off ice, and cam e in.
Dunster saw that Eberhard's complexion was a
Ii ttle ruddier than usual, and the r e 1'1 a s
no smile on his face. Dunster concluded that
Eberhard had had a couple of drinks, perhaps
more.
"Andy, old pal," s aid Eberhard, "this j at
mess has got to stop." But the tone MacNei 1
listened to was even less friendly than the
l"lords.
"Harry, old pal," said Dunster even I y
"you can't l\fant that any worse than I do."
"Well, 111hy don't you stop it?"

- 25 -

AUTOMATIC \fIXUP

"As a matter of fact," said Dunster, "i t probably has stopped -- for now."
"What," said Eberhard, "you lOOan the troub:E
in Natrasys has stopped?"
"Yes, " said Dunster, smiling, "the f requency of complaints t"8S slowing down from 1600
to 1630, and after 1630 I think only 0 ne has
come in."
"Well," said Eberhard, "so you fixed the
trouble," and he began to relax and smile.
"I

! "No, Ive didn't detect the trouble. We had

all our inspection crews testing the equipment,
and t~en the trouble just ceased, died a way.
Furthe1rmore, "since the trouble is no Ion g e r
manifes ting itself, I doubt that we can f iii d
it by testing, for all the equipment is again
functioning perfectly."
"Yes," said Eberhard reflectively, "I can
see that, but I feel as if I were in a nightmare. "
"We can how'ever examine and analyse the
383 complaints, and see what sort of thing they
indicate."

, "No, not really, because Crurmvell gave us
the start of Natrasys," said Dunster.
"You
can't make great advances without money to invest -- at least you can't often. Crumwell
is still supplying most' of our working capitaL"
Eberhard got up, paced to and fro in the
room, and tvrung his hands. Dunster watc h e d
him. Then he got up, put his ar,m around Eberhard's shoulders, and said, "Si t d01~n, Harry;
take it easy; we're a long way from being
licked."
Eberhard sat down, stretched himself out,
held his head in his hands as if it were aching
and sagged like a pillow with no feathers.
Jim Halvker came back into the room. "Professor, here is your analysis. They promised,
it to me every felf minutes, and so I w ai ted.
But Josophat, Professor, the tdres, the co mplaints, have stopped coming in. Wha'd'yo u
know?"
Dunster reached for the analysis, too k
it, sat dOtvn at his desk, and Ivas los t in concentration.
Jim Hawker said, "Professor, excuse me,
I might as well let the inspection crelV s go
home, I suppose, it's 1820?"

"Well, it's your responsibility, Andy."
"You're right, Harry" said Dunster deliberately and looking at Eberhard intently.
Eberhard went on, "CrulJlfell was bothering
me the other day. He said he Ivas tired of mmkeying around Id th us. We had rolled up a fine
record, he said sarcastically, of eleven pieces
of trouble in six years. We had lost something
like a quarter of a million dollars as a result,
he said, and he Ivas tired, jat tired of us. He
s aid one more piece of trouble and hew a s
through, j at it, through." Eberhard add ed,
"I hope CrufIl.\,ell doesn't hear of the trouble
today."
Dunster said: "But he's bound to, Harry.
This is the-l-vorst piece of trouble we've ever
had. I think we'll be sued for something like
six million dollars, estimating an average 0 f
$20,000 damage for each of 300 cases."
"Oh, jat, jat, jat, jat," said Eberhard
feebly.
Dunster smiled again, "Harry, you may as
l\'ell set to lwrk thinking of a better answer
than that for your friend Crulllvell when he gets
on the phone tomorrow about this."
"My friend!" said Eberhard, "the hell you
say. lIe's no friend of mine. IldshI'dnever
seen him."
- 26 -

Dunster looked up long enough t 0 say:
"Sure, Jim. But have one cretv stay in t'\fatc h
quarters -- you can never tell when our Josophat Bug lvill start gnatving away again."
Dunster found the analysis most interesting.
Almost all the complaints were misdirections, as if the clerks who had entered the
original shipping instructions had chose nat
random a wrong destination for an othe r w is e
correct shipment. It was as if some drug had
affected them causing momentary loss of co nsciousness. But how could any drug a f f e c t
so many clerks at so many shipping offices?
He looked through the analysis she e t s
for the table that showed the geogr a ph i cal
dis tribution of complaints according to shipping
office. There Ivas, it seemed, a definite tendency for complaints to concentrate accordin g
to size of shipping office on the f 0 11 ow ing
principle: the larger the volume of Shipments
from a shipping office, the greater the number
of complaints. Another look shOtved that the
shipping offices connected tvi th the Sec 0 n d
Central Office of Natrasys, on the West Coast
in Netv Frisco, had produced no complaints. The
equipment there had not been affected by the
trouble. The trouble tvas definitely ass ociated with the first Central Office, Net'\' Chicago.

AUTOMATIC MIXUP

The analysis also showed that all the de":'
fee ti ve shipments originated in the per i 0 d
1012 to 1547 on 10 September.
Other than these indic ations, Dun s t e r
fel t that the analysis did not Sh01i m u c h ,
aside from the fact that his earlier estimate
of loss was too high. The average claim fo r
damage was going to be far less than $20,000.
In fact, he would estimate it at 300 tim e s
$5,000, or one and a half millions. Even so,
Mr. Valerian Crurmiell was definitely not going
to be pleased.
Dunster leaned back in his chair, turning
possibi Ii ties over in h is head. Eber hard's
eyes were shut and his mouth was open; he lias
making little snoring sounds.
The next day Natrasys had no t r 0 ubI e •
Due to some kind of miracle, l'llhich Dunster did
not understand -- and which he felt he had no
basis yet for understanding -- the equipment
of Natrasys apparently lVorked perfectly.

But the third day, around 0940, as D u nster came into the Central Control room, from
the corridor that connected with the Computation Laboratory, Jim Hawker came up to him and
said, "This just came in, Professor," Dunster
read:
CENTRAL CONTROL, NATRASYS, NEW CHICAm
-- FARMER G. OTIS HERE CLAIMING $200,000
DAMAGES FOR 8,000 CHINCHILLIZARD HJUaUNG
EGGS DELIVERED DEAD. SHIPMENT SCHEDULED
EXPRESS DETROIT TO KALAMAZOO, LOST, FOUND
AT MIAMI, RESHIPPED. WHY TIUS ERRO R ?
PLEASE ADVISE. URGENT.
-- I. R. CONE, KALAMAZOO ,
0920, 13 SfllTEMBER
"Crurmvell will sure throw a fi t if he hears
about this," said Hatvker.
"You are right, Jim," said Dunster. "But
the tdre does not say the time when this shipment left Detroit. Can you find that out?"
Hatvker said, "Yes, I will," and started
to ualk d01'\1n to the Analysis Section. But a
minute later Hal~ker came back tvi th t t'l 0 more
telegrams. "IT, the Josophat IT," he said and
paused, "has started again."
Duns ter took the wires and read the m :
"Yes, Jim. The Josophat Bug -- Get the teste rehls going again, Jim. Have them concentrate
on the condition of the machine at the tim e
hlhen the faulty shipments took place. Or ask
them to if they can
it is an enormous number of microseconds in the past."

- 27 -

IV.

Valerian Crurmvell was alone, and tvalking
back and forth on the thick soft dull-red carpet in his office on the 45th floor of the
Board of Commerce Building. It was a big room.
His des k, a great wa lnu t aff air, was i n the
center of the room; over by one wall tV a s a
conference table large enough for a dozen men.
There were Hind01'/s on three sides of the room.
Crurmvell rather liked his room as a rule. But
today, 13 September, he t'/as in no mood to enjoy
it. He had been out of the office yesterda~
to play golf. But he had .made a mi s t a k e in
choosing a partner, and he had been be a ten.
He did not enjoy being beaten. He was deeply
irritated.
His secretary opened the door and shrunk
a little, trying to make herself smaller an d
less conspicuous. "Mr. Crurm'Vell," she sa i d
timidly, "there is a long distance call for
you from Kalamazoo, a Mr. Otis - he said i t
lias important -- he said he had to speak to you
personally."
Crurmvell said in his loud voice, "y 0 u
told him I was in conference?" The l i t tIe
secretary tried to shrink some more, andsaid,
"Yes, but he insisted it was urgent, that I
interrupt you -- I did not want to do it, Mr.
Crurmvell. He said it was about your horrible
National Traffic Management System and 8000
murders.
"What?" said CruI1l'Jell, "what -- 8000 murders? Give me the phone." But he reach e d
over to the phone on his desk and pic ked it
up. "Hello," he practically shouted into the
mouthpiece, "this is CruI1l'l1ell."
"Mr. Crurmvell, " said a thin high-pitched
bi ting voice on the other end of the lin e,
"Mr. Crurmvell, your horrible National Traffic
Management System has jus t murdered 8000 -just think, 8000 -- of my precious chinch i 1lizard eggs t'Vorth $25 apiece."
"What?" roared CruIm\lell, "how's that?"
"Mr. Crurmrell, they were shipped yesterday
from Detroit addressed to me. It takes hal f
an hour by special delivery missile.
Chi nchillizard hatching eggs can stand motion and
coolness for about two hours and then they die.
Well, {vhat do you suppose your horrible system
did for me?"
"What?" shouted CruI1lvell.
"Well those 8000 precious eggs 1.lIen t to
Miami, Florida, -- just think -- Miami instead
of Kalamazoo, and got lost down there, and did

AUTOMATIC MIXUP
not get to me until about 0800 this morning,
and every single one of my poor little chi nchillizard eggs is dead." The voice waile d
at the other end of the phone, and then itbecame a shout: "And I shall have you knOt', Mr.
Crumwell, I shall SUE YOU FOR $200,000. I am
going to hire the best lawyers in the Midwest,
and George Otis of Kalamazoo will beat Valerian CrulJloiell of New Chicago."
"Mr. Otis," Crurmvell shouted, "1 e t me
look into this, I'll call you back in an hour
or two, just as soon as I can get to the bottom of this."
The voice said, "Do as you like, Mr. CrwnYou are going to get a summons in just
about three hours. And you are going to pay
me $200,000."

t

- 33

,

'

,

"s

.."..,

'

ASSOCIA TION FOR COMPUTING MACHINERY MEETING,
PHILADELPHIA, SEPT. 14

to

16, 1955

Names and Addresses of Authors of Papers

In the November, 1955, issue of "Computers ani
Automation", we printed the titles, authors, and
abstracts of the papers given at the meeting of llie
Association for Computing Machinery at the University of Pennsylvania, Philadelphia, Pa., Sept. 14
to 16, 1955. These titles and abstracts were rep'roduced photographically from the program distributed at the meeting; since the program did not
give the authors' addresses, they t...ere not printe d.
This lack is nOI\' remedied. Following is the
list of the names and addresses of authors, kindly
furnished by Professor John P. Nash, Digital Computer Laboratory, University of Illinois, Urbana,
Ill., chairman of the Program Committee of the Association for Computing Machinery. The numbers below agree with the numbers of the papers shown on
pages 17,to 32 of the November issue.
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.

David P. Perry, Sperry Rand Corp., 315 Fourth
Ave., New York, N.Y.
Mark Lotkin, RCA Service Co., Inc., Missile
Test Project, Patrick AFB, Fla.
Richard E. von IIoldt, Radiation Laboratory,
Uni v. of Calif., Berkeley 4, Calif.
Wallace Givens, Dept. of Math., Univ. of Thnn.
Knoxville, Tenn., or Oak Ridge Nat'l Lab., Oak
Ridge, Tenn.
Rob ert Perkins, Ramo-Wooldridge Corp., 5740
Arbor Vitae St., Los Angeles 45, Calif.
Hollis A. Kinslow, IBM Corporation, P.o. Box
390, Poughkeepsie, New York
E. K. Blum, U.S. Naval Ordnance Lab., Whi te
Oak, Silver Spring, Md.
J. N. P. Hume, Computation Centre, McLennan
Lab., Univ. of Toronto, Toronto, Canada
Charles E. Thompson, General Electric Corp.,
Richl3nd, Wash.
John R. Stock, United Carbide and Carbon Corp.,
30 E~ 42nd St., New York 17, N.Y.
Stanley Frankel (Continental Oil Co.) 135 Main
St., Seal Beach, Calif.
G. M. Amdahl and J.W. Bachus, IBM Corp., P. O.
Box 390, Poughkeepsie, N.Y.
William E. Smith, North American Aviation Co~,
12214 Lakell'ood Boulevard, Downey, Calif.
Ch'arles J. Sldft, CONVAIR, San Diego, Calif.
W. R. Hoover, Calif. Inst. of Tech., Pasadena,
Calif.
Norman Grieser, Underwood Corp., Electronic
Computer Div., 35-10 36th Ave., L.I.C. 6, N.Y.
J. A. Postley, The Rand Corp., 1700 N. Main
St., Santa Monica, Calif.
L. Eselsen', Remington Rand, Inc., 315 Fourth
Ave., New York 10, N.Y.
Mary K. Hawes, Sperry Rand Corp., 315 Fourth
Ave., New York 10, N.Y.
Leon Nemerever, Unde~iood Corp., Electronic
Computer Div., 35-10 36th Ave., L.I.C. ,6, N.Y.
J. A. Porter and D. L. Shell, General Electric
Co., Computer Techniques Development, Investigation Sect., AGT Development Dept., Bldg.
300, Cincinnati 15, o.
- 34 -

22. A. E. Roberts, Jr., General Kinetics, Inc •• 555
23rd St., S., Arlington 2, Va.
23. Dean Arden, Mass. Inst. of Tech., Digital Computer Lab., Cambridge, Mass.
24. James E. Robertson, Digital Computer Lab., 168
Eng. Res. Lab., Univ. of Illinois, Urbana, Ill.
25. Norman R. Scott, Dept. of Electrical Engineering., Univ. of Michigan, Ann Arbor, Mich.
26. Earl J. Isaac and Richard C. Singleton, Stanford Research Inst., Stanford, Calif.
27. Benjamin L. Schwartz, Battelle Memorial Inst.,
Cleveland, o.
28. R. L. Cline, IBM Corp., P. O. Box 390, Poughkeepsie, N. Y.
29. Edward II. Friend, N. Y. Life Insurance Co., 51
Madison Ave., New York 10, N.Y.
30. J. H. Allen, Temco Aircraft Corp., P. O. Box
6191, Engineering Dept., Dallas, Texas
31. A. S. Householder, Math. Div., Oak Ridge Nat~
Lab., P. O. Box P, Oak Ridge, Tenn.
32. J. H. Wegstein, Nat'l Bureau of Standards, Wadl.,
25, D. C.
33. S. G. Campbell, Oak Ridge Nat'l Lab., P. O.Bo~
P., Oak Ridge, Tenn.
34. Norman F. Loretz, Magndvox Research Lab., Van
Nuys, Calif.
35. Alan J. Perlis, Head, Computation Lab., Purdue
Univ., LaLayette, Indiana
36. C. W. Adams, Office Methods and Procc d ures
Westinghouse Electric Corp., Pittsburgh, Pat
37. E. M. Gettel and D. L. Shell, Computer Te ch niques Development, Investigations Sect., AGT
Development Dept., Bldg 300, General Electric
Co., Cincinnati 15, o.
38. Stefan Bergman, Stanford Univ., Applied Math.
and Statistics Lab., Stanford, Calif.
39. Stanley Katz and Jack Warga, Elec troData Corp.,
460 Sierra Madre Villa, Pasadena, Calif.
40. Jack Beller, H. B. Keller, Samuel Schechter,
Inst. of Math Sciences, AEC Computing Facility, 253 Greene St., New York 3, N.Y.
41. Barry Gordon, Equitable Life Assurance Society,
393 Seventh Ave., New York I, N.Y.
42. Walter F. Bauer, Ramo-Wooldridge Corp., 5740
Arbor Vitae St., Los Angeles 45, Calif.
43. J. Wright, Engineering Research Inst., Willow
Run Research Center, Univ. of Michigan, Ypsilanti, Mich.
44. J. H. Brown, John W. Carr III, Boyd Larrowe,
J. R. McReynolds, Engin~ering Research Inst.,
Willow Run Research Center, Univ. Jf Michigan,
Ypsilanti, Mich.
45. F. S. Beckman and D. A. Quarles, Jr., IBMCo~
590 Madison Ave., New York, N.Y.
46. David J. Fitch, Student Counseling Bureau,
Univ. of Illinois, Urbana, Illinois
47. William Orchard-Hays, The Rand Corp., 1700
Main St., Santa Monica, Calif.
48. Joseph V. Natrella, Directorate of Management
Analysis, IIdqtrs., U.S.A.F., Washington, D.C.
49. John Greenstadt, IBM Corp., 590 Madison Ave.,
Net~ York, N.Y.
50. E. G. Kogbetliantz (paper cancelled),IBM Corp~

ACM MEETING
51.
52.
53.
54.

55.
56.
57.
58.
59.
60.
61.
62.
63.
64.
65.
66.
67.
68.
69.
70.
71.
72.
73.

74.

75.
76.
77.
78.

P. O. Box 390, Poughkeepsie, N.Y.
Arthur G. Dmming and Alston S. Householder,
Oak Ridge Nat' 1 Lab., P.O. Box P, Oak Ridge,
Tenn.
Wnllace Givens, Univ. of Tennessee, Dept. of
Math., Knoxville, Tenn., or Oak Ridge Nat'l
Lab., O~k Ridge, Tenn.
Stephen E. Wright, Sperry Rand Corp., 315
Fourth Ave., New York 10, N. Y.
Bruce G. Oldfield and Robert H. Bracken, U.S.
Naval Ordnance Test Station, China Lake,
Calif •
Bruce G. Oldfield, U.S. Naval Ordnance Test
Station, China Lake, Calif.
Howard S. Levin, Ebasco Services, Inc., 2
Rector St., New York 6, N.Y.
IIe len ~leek and Leon Gainen, Hughes Research
and Development Labs., Hughes Aircraft Co.,
Culver City, Calif.
Ned Chapin, Illinois Inst. of Tech., 407 Gunsaulus Hall, 3140 S. Michigan Ave., Chicago
. 16, Illinois.
R. J. Rossheim, U.S.Steel Corp., 525 William·
Penn Place, Pittsburgh 30, Pa.
G. E. Forsythe, Numerical Analysis Research
Univ. of Calif., Los Angeles, Calif.
A. B. Taub, Digital Computer Lab., 168 Eng.
Res. Lab., Univ of Illinois, Urbana, Ill.
Wesley S. Melahn, The Rand Corp., 1700 N.
Main St., Santa Monica, Calif.
Charles L. Baker, Douglas Aircraft Co. ,Santa
Monica, Calif.
Ot~en R. Mock, North American Aviation, 12214
Lakewood Boulevard, Downey, Calif.
Ramon Alonso and Thomas Conley, Ballistic ~
search tabs., Computing Lab., Aberdeen Proving Ground, Aberdeen, Md.
Bill L. Wadell, G. M. Giannini and Co., 918
E. Green St., Pasadena 1, Calif.
R. E. Me I'\·dn, IBM Corp., P. O. Box 390,Poughkeepsie, N.Y.
Raymond Stuart-Williams, Mil ton Rosenberg
and M. A. Alexander, Internat'l Telemeter
Corp., 2000 Stoner Ave., Los Angeles, Calif.
John Todd and Phillip Davis, Computation Gp.,
Nat'l Bureau of Standards, Washington 25,D.C.
J. Pasta and S. Ulam, Computer Lab., Los Alamos Scientific Labs., P. O. Box 1663, Los
AlnmtJs, New Mexico
Robert C. Miller, Jr., and Eruce G. Oldfiel~
Naval Ordnance Test Station, China Lake,Calif.
Jules I. Schwartz and Gus S. Hempstead, The
Rand Corp., 1700 N. Main St., Santa Monica,
Calif •
Richard C. Luke, Lockheed Aircraft Corp., Van
Nuys, Calif.
John I. Derr, The Rand Corp., 1700 N. Main
St., Santa Monica, Calif.
Hiram G. Martin, Douglas Aircraft Co. ,Sa n ta
Monica, Calif.
II'\dn D. ~reenl"ald, Rand Corp., 1700 N. ~tIin
St., Santa Monica, Calif.
.
R. K. Gerlach and D. O. Miles, Rand Corp.,
1700 N. Main St., Santa Monica, Calif.
W. J. Poppelbaum, Digi tal Computer Lab., 168
Eng. Res. Lab., Univ. of Illinois, Urbana, Ill.
John R. Bethke, Bur~oughs Corp., Research
Center, Paoli, Pa.
E. R. Beck, Bendix Aviation Corp., Research
Labs Di v., 4855 Fourth Ave., Detroit, ~1ich.

. 79.
80.

81.
82.
83.

84.
85.
86.
87.

88.
89.
90.

91.

J. L. Smith and A. Weinberger, Nat'l Bureau
of Standards, Washington 25, D. C.
E. G. Kogbetliantz, IBM Corp., P. O. Box 3~,
Poughkeepsie, N. Y.
Peter Henrici, Nat'l Bureau of Standards,
Washington 25, D. C.
R. W. Bemer, Lockheed Aircraft Corp., Van
Nuys, Calif.
Bengt Carlson and Max Goldstein, Los Alamos
Lab., P. O. Box 1663, Los Alamos, New Mex.
Saul Gorn, Aberdeen Proving Ground, Computing Lab., Md.
Franz Hahn, 56 Electrical Engineering Bldg.
Univ. of Illinois, Urbana, Ill.
Arthur E. Hoerl, E. 1. DuPont de Nemours Co.
Experim~ntal Station, Wilmington 98, Dela.
William Miehle, Burroughs Research Center,
Paoli, Pa.
James L. Maddox and Ralph H. Beter, Philco
Corp., Philadelphia 44, Pa.
Richard C. Jeffrey, MIT, Cambridge, Mass.
Max A. Woodbury, George Washington Univ.,
Washington, D. C.
Gene H. Golub, Univ. of Illinois, Digital
Computer Lab., Eng. Res. Lnb. (68), Urbana,
Ill.

92.
93.
94.

95.
96.
97.

98.

Arthur E. Hoerl, E. 1. DuPont de Nemours Co.,
Experimental Station, Wilmington 98, Dela.
Robert II. Braken, U.S. Naval Ordnance Test
Station, China Lake, Calif.
I. McNamee and E. D. Fullenwider, U. S. Naval Ordnance Test Station, Corona, Calif.
H. Flatt, Ballistic Research Labs., Aberdeen
Proving Ground, Computing Lab., Aberdeen. ~.:d.
L C. Long, Oak Ridge Nat'l Lab., P. O. BoxP
Oak Ridge, Tenn.
Charles W. Adams, Westinghouse Electric Corp.,
Pittsburgh, Pa.
David E. Muller, Digital Computer Lab., 168
Eng. Res. Lab., Univ. of Illinois, Urbana,
Ill.

99.
100.

D. A. Huffman, MIT, Cambridge, Mass.
Manfred Kochen, Ins t. for Advanced Stu d y,
Princeton, N. J.
101. John O. Lilly, U.S. Naval Ordnance Test Station, China Lake. Calif.
102. James L. McPherson, Bureau of the Census,
Dept. of Commerce, Washington 25, D. C.
103. W. C. Jacob, Univ. of Illinois, Dept.of Agricul ture, 208b Davenport Ball, Urbana, TIL
104. Christine Kris, Univ., of Chicago, 6109 S.
Ellis Ave., Chicago 37, Ill.
105. Jack C. Me I'\d n , Syracuse Univ., Syracuse, N.Y.
106. C. L. Gerberich and W. C. Sangren, Oak Ridge
Nat'l Labs, P. O. Box P, Oak Ridge, Tenn.
107. R. J. Arms, D. F. Eliezer, L. O. Gates, Jr.
an~ D. ~. Young, Jr., U.S. Naval Proving
Ground, Dahlgren, Va.
108. Elizabeth Cuthill and Ruth M. Davis, David
Tayl~r Model Basin, Washington 7, D. C.
109. Robert C. Miller, Jr. and Ralph G. Selfridge
U. S. Naval Ordnance Test Station, China
Lake, Calif.
110 •. W. M. Harris, A. O. Smith Corp., Milwaukee
1, Wisc.
111. P. A. Zaphyr, W~stinghouse Electric Corp.,
East Pittsburgh, Pa.
112. Stephen E. Wright, Sperry Rand Corp., 315
Fourth Avenue, New York 10, N.Y.
113. Nyles V. Reinfeld (paper cancelled) Exreutive
(continued on page 37)

- 35 -

MAN

U

S

We are interested in articles, papers, reference information, science fiction, and discussion relating to computers and automation.
To be considered for any particular issue, the
manuscript should be in our hands by the fifth
of the preceding month.
Articles. We desi re to publish articles t hat
are factual, useful, understandable, and interesting to many kinds of people engaged in one
part or another of the field of computers and
automation. In this audience are many people
Ivho have expert knOt~ledge of some part of the
field, but Ivho are laymen in other parts of it.
Consequently a writer should seek to explai n
his subject, and show its context and significance. He should define unfamiliar terms, or
use them in a way that makes their meaning unmistakable. lIe should identify un fa mil i a r
persons lvi th a feH IVOrdS. He should use examples, details, comparisons, analogies, etc.,
whenever they may help readers to understand
a difficult point. He should give data s upporting his argument and evidence for his assertions. We look particularly for articles that
explore ideas in the field of computers and
automation, and their applications and i mpli-

c

R

I

P

T

S

cations. An article may certainly be co n troversial if the subject is discussed reasonabl~
Ordinarily, the length should be 1000 to 4000
words. A suggestion for an article should be
submitted to us before too much work is done.
Technical Papers. Many of the foregoing r equirements for articles do not n e c e s sa r i 1 y
apply to technical papers. Undefined technical terms, unfamiliar assumptions, mathematics,
circuit diagrams, etc., may be entirely appropriate. Topics interesting probably to only
a few people are acceptable.
Reference Information. We desire to print or
reprint reference information: lists, ros~r~
abstracts, bibliographies, etc., of use to computer people. We are interested in ma kin g
arrangements for systematic publication from
time to time of such information, with other
people besides our own staff., A nJ. 0 new h 0
would like to take the responsibility for a
type of reference information should write us.
Fiction. We desire to print or reprint fiction
which explores scientific ideas and possibilities Qbout computing machinery, robots, cyber(continued on page 37)

*--------------------------------------------- * ------------------------------

- 36 -

""

Ar.\l ~1F.F.TI~r.
(contInued from,.p.age 35)
Services, Cleveland lu: o.
114. Arnold Siegel, MIT, Cambridge, Mass.
115. R. E. Spero and J. Stuart, Sperry Gyroscope
Co., Creat Neck, N.Y.
116. J. K. Slap, Northrop Aircraft Corp., Hawthorne, Calif.
117. A. Karen and B. Loveman, Reeves Instrument
Co., 215 E. 9lst St., New York 28, N.Y.
118. Nathaniel Macon, Alabama Polytechnic Inst.,
Auburn, Ala.
119. R. S. Lehman and G. II. Weiss, Ballistic Research Labs., Aberdeen Proving Ground, Md.
120 • W. F. Atchison, Georgia Inst. of Tech. Rich
. Computing Center, Atlanta, Ga.
F. T. !Vall and L. A. lIiller, Jr., Dept.of
CI,cmistry, Univ. of Illinois, Urbana, Ill.
121. J. Gardner, R. Page and O. Tiffan, Willow
Run Research Center, Univ. of Michigan, 'lpsi-

THE COMPUTER DIRECTORY, 1956:
NOTICE

The June 1956 issue of "Computers and Automation" will be the second issue of "The Computer Direc tory". Las t year tie published the
first issue, 164 pages. Our present plans for
the June 1956 directory follow:
Part 1 of the directory in 1956 will be a
cumulative "Roster of Organizations in the Computer Field" based on the last cumulative roster (published December 1955, containing about
330 entries) and brought up to date. Entries
in this roster will be free. If you know 0 f
any changes, addi tions, or correc tions ti hi c h
should be made in the' entries, please tell u s.

l~mti, ~lich.

122.
123.
124.
125.
126.
127.
128.
129.

R. E. Utman and Margaret II. lIarper, Remington Rand, Inc., 1624 Locust St., Philadelphia, Pa.
Anthony G. Oettinger, Computation Lab., !Iarvnr I TJni v., Cambridge 38, Mass.
W. F. Bauer and A. Vazsonyi, Ramo-Woohlridge
Corp., 5740 Arbor Vitae St., Los Ange~es 45,
Calif •
R. P. Beals, Chrys ler Corp., Parts Di v., 7000
E. Eleven Mile Rd., Center Line, Mich.
Louis B. Wadel and B. B. Mackey, Chance Vou91t
Aire"t··1f t." Tnc., P. O. Box 5907, Dallas, Tex.
Takeo Miura and Toshiro Numakura, Hitachi
Central Research, Kokubunju, Tokyo, Japan
L. E. Heitzer, CONVAIR, Fort Worth, Texas.
M. Yanowitch, Reeves Instrument Co., 215 E.
9lst St., New York 28, N.Y.

Part 2 of the directory will be the secorrl
edition of "The Computing Machinery Field: Produc ts and Services for Sale." Over 600 entries
on 21 'pages appeared in the first edi tion i n
June 195;1; a considerable increase is anticipated. The previous entries, and blank forms\
will be'sent in February, to suppliers for review, checking, and additions. It i~ expected
at this time that a nominal charge of $6.00 an
entry will be requested from each supplier in
order to help defray the cost of preparing and
printing the directory; but if the charge i s
not paid, t:1e entry may still appear in condensed form, if desirable to make the listi n g
complete.

_ ENTl -

:;.----------

~ANusrRIPT

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

Part 3 of the directory will be the third
edition of the Who's Who in. the Computer Field
In the June 1955 issue, about 7500 entries appeared on 96 pages; of these about 2600 l'J ere
full entries, and the remainder were brief entries. Our present plans are to publish only
new or revised Who's Who information in the
June 1956 directory. Blank forms for new 0 r
revised entries will be sent in Feb ru a ry 0 r
March to all computer pe a pIe lve know of. It
is expected at this time that a nominal charge
of $2.00 an entry will be requested from each
person whose entry is printed, in order tohe~
defray the cost of preparing and printing the
Who's Who; but if the charge is not paid, a
b~ief entry may ~p~ear in condensed form if
desirable to make the listing complete.

NOTICE

(continued from page

36)

netics, automation, etc., and their implications,
nnd which at the same time is a good story.
Ordinarily, the length should be 1000 to 4000
l'Jords.
Discuss ion. We des ire to print in "F 0 rum"
brief discussions, arguments, announcements,
nelVS, letters, descriptions of remarkable nel'V
developments, etc., anything likely to be 0 f
substantial interest to computer people.

The main reason for the nominal charges
mentioned above is that we look on the directory as a service to m~ny people in the computer field; yet so far it has not paid for itself; and we need to make a compromise, p u blishing at least some information about everything that should appear in the directory, but
fuller information for those who have shared
directly in the cost.

Payments. In many cases, l'Je make small token
payments for articles, papers, and fiction, if
the author lvishes to be paid. The rat e i s
ordinarly !i¢ a word, the maximum is $20, and
both depend on length in l'wrds, whether printed
before, whether article or paper, etc.

- 37 -

SP~CIAL
~~C.OMPUTERS_

ISSUES OF
AND AUTOMATIO~"

BULK SUBSCRIPTION RATES
These rates apply to subscriptions coming in
together direct to the publisher. For example,
if 5 subscriptions come in together, the saving on each one-year subscription will be 24
percent, and on each two-year subscription will
he 31 percent. The bulk subscription rate s,
depending on the number of simul taneous s u bscriptions received, follow:

The June issue of "Computers and Automation" commencing with June, 1955, is aspecial
issue, "The Computer Directory."
For details about the next compu te r
directory, see "The Computer Directory. 1956:
Notice."

Bulk Subscription Rates
(United States)

* - - - - - - - - - ~; - - - - - - - - - *,

Rate for Each Subscription, and
Number of
Sirnul taneous Resulting Saving to Subscriber
Subscriptions
One Year
Two Year
10 or more
5 to 9
4
3
2

PATENTS
(continued from page 32)

and Richard H. Richwine, Greenfield, Indiana
/ - / An undirectional, one;~ay automatic
spring clutch.
2,723,080 / Daniel L. Curtis, Venfse, Calif. /
Hughes Aircraft Company / A device, in co mbination t'li th a bistable flip-flop, for rendering said flip-flop responsive to one 0 f
two input pulses applied simul taneously trereto.
2,723,081 / Harry B. Miller, Wartdck, R. 1. /
Max L. Grant, Providence, R. I. / An overrun
control for rotary-counter registers.
2,723,312 / John H. McGuigan, Net~ Providence.
N. J., and Orlando J. Murphy, New York, N.Y.
/ Bell Telephone Laboratories, Inc., Netl' YolK,
N·. Y. / A magnetic drum dial pulse recording
and storage registers.
2,723,347 / Louis F. Mayle, Fort Wayne, Ind. /
Farnsworth Research Corporation, Fort Wayne,
Ind. / A pulse keying circuit for a p Ot\1 er
amplifier.
2,723,352 / Willia~ C. Sealey, Wauwatosa, Wisc.
/ Allis-Chalmers Manufacturing Co., Milt\'aukee,
Wisconsin / A polyphase regulating system for
obtai~ing balanced voltag~s.
2,723,353 / Charles F. Spitzer, Syracuse, and
Robert T. Gordon, Camillus, N.Y. / General
Electric Co. / A non-linear resonant trigger
circuit.
2,723,355 / Robert E. Graham, Chathorn Tow n ship, Morris County, N.J. / Bell Telephone
Laboratories, Inc., New York, N.Y. / A diode
gate circuit.
2,723,365 / Charles R. Williams, IIawtharne,Calif.
/ Northrop Aircraft, Inc., H31'lthorne, Calif./
A sectional read-out tube and circuit.
- END -

- 38 -

$3.80, 31%
.4.20, 24
4.60, 16
5.00, 9
5.25, 5

$6.60,
7.25,
8.00,
8.80,
9.55,

3-ro,;6
31
24
16
9

For Canada, add 50 cents for each year; 0 u tside of the United States and Canada, add $1.00
for each year.

COM P U T,E R

Applications
and

Development
Computing Services
Engineering and scientific computation
Analog Digital

Computers Systems Development
Analog and Digital
Systems analysis and synthesis
Logical organization
Circuit design
These are full time positions offering fascinating
and varied work and educational benefits for
graduate study. Contact:

J. A. Metzger
ARMOUR RESEARCH FOUNDATION

of Illinois Institute of Technology
10 West 35th Street
Chicago 16. Illinois

FERRANTI

PUBLICATIONS

p 34; LINEAr. pnOGrA~l~lING
AND CO~]PtJTEnS.
Reprint of tt"o

HIGH SPEED 'APE READER

articles by Chandler Davis, in July
and August 1955 "Computers and Automation". A clear, t'iell-wri tten introduction to linear programming,wiili
emphasis on the ideas.
• ••• $1.20
P 20: THE COMPUTER DIRECTORY, 1955.
164 pages, 7500 Who's Who entries,
300 Organization entries, and 600
entries of Products and Services for
Sale in the Computer Field; 250,000'
words of condensed factual information about the computer field, June
1955 issue of "Computers and Automation. "
•••• $4.00

('

P 32: SYMBOLIC LOGIC, by LEWIS CARROLL. Reprint of "Symbolic Log i c,
Part I, Elementary," 4th editicn, 1897,
240 pages, by Lewis Carioll (C. L.
Dodgson). Contains Lewis Carroll's
inimi table and entertaining problems
in symbolic logic, his method of solution (now partly out of date), and
h is sketches of Parts II and III,
Hhich he never wrote since he die d
in 1898.
• ••• $2.50
P 25: NlJMBLES
NUMBER PUZZLES FOR
NIMBLE MINDS. Report. Con t a ins
collection of puzzles like:

I

I

TRY II A V E and T R A I N
S E
FUN your WIT S
=TWVAS WASE
ENTNS

+ TIlE

WYE

=VIF

In fact, you can also: 90893 85202
44393 29081 (Solve for the digits-each letter stands for just one digit
o to 9)
All are new numbles, additions, multiplications, etc.; some easy, some
hard. Each wi th th'O messages, 0 n e
open, one hidden. HInts for solutIon.
Good exercises in logical reasoning.
•• • •$1.00

.'

--------MAIL TIllS COUPON---------or copy it
Edmund C. Berkeley and Associates,
815 Washington St., R142,
Newtonville 60, Mass.
Please send me publications circled
and your announcement of publications:
2D

25

32

34

"

.:c ::" (

The Ferranti High Speed Tape Reader accelerates
to full speed within 5 milliseconds and stops within,
3 milliseconds. It has been in use at leading computer
installations for over two years and has achieved a sound
reputation for simplicity and reliability in regular operat!on.

-I-\i' (

I) Mark II model reads at speeds up to 200 characters per
second, and stops the tape from full' speed within a character positionwithin .03 inch: The tape is accelerated to full speed again in 5 milliseconds
and the following character is ready for reading within 6 milliseconds of
rest position.
(2) Mark IIA model reads at speeds up to 400 characters per second,
and stops within .1 inch.

'1#;£!-",,1

Both models read either 5 level, 6 level or 7 level tape
by simple adjustment of an external lever.
WiMt4!WThe tape is easily inserted without complicated threading.
Lap or butt splices are taken without ally difficulty. The same tape may be
passed' thousands of times without appreciable tape wear. The optical
system has no lenses or mirrors to get out of alignment. Friction drive is
independent of sprocket hole spacing.

LARGE OUTPUT Amplifiers are included for each channel, including
a special squaring circuit for the sprocket hole signal. Output swing
between hole and blank is greater than 20 volts.
Dimensions: 91( x 11 V2" x 111,4"
Weight: 37 Ibs.
For use with long length~ 6f tape up to 1000 feet, spooling

equipment operating up to 40 inches per second for take-up
or supply is available separately.

I enclose $
in full payment.
(Add 10¢ per i tern to cover cos t 0 f
handling and mailing.) It is understood that if I am not satisfied, I
may re t urn any i tern tV i th in five days
after receiving it, and you will at
once refund my money. My name and
address are attached.

FERRANTI ELECTRIC, INC.
30 Rock.f.lI.r Plaza

- 39 -

New York 20, N. Y.

Forum
ANOTHER PROGRAMMING FAILURE
Bill Danch
Munich, Germany

"I'll have to get a new computer cook book. This upside
down cake recipe is coming out right side up again:"
- tiD -

MAKE YOUR OWN BABY GENIUS COMPUTERS
WITH

GENIAC
Electric
Construction

Brain
Kit

No.1

Diagram of the versa• ~ : : .' tile multiple switch,
'. " ~ " .' .\ which can be assembled
.':::' ......: ••., to make any s wit c h
combinations from 16
, ••••.:. : "':".: . decks of 2 positions,
~
10 decks of 3 p os itions, etc., to 2 decks
of 16 positions.

.'. :. 'J'
Q3

This ki t is an introduc tion to the
design of arithmetical, logica 1 ,
reas oning, comru ting, puzz le-s 01 viJJ],
and game-playing circuits. It i s
simple enough for intelligent boy s
to assemble, and yet is instructive
to computer men because it ShOHS how
many kinds of computing and reasoning circuits can be made fromsimple
components.
Wi th this ki t and 64-page ma n u aI,
you can eas ily make over 30 sma 1 1
elec tric brain machines that exhibit
intelligent behavior. Each runs on
one flashlight battery. All connections uith nuts and bolts; no s 0 1dering required. Price, $17.95 (add
80t for shipment in U. S. \'Ves t of
Mississippi, $1.80 for s hip men t
outside U. S.). If not satisfactory,
returnable in seven days for full
refund.
A few of the machines you can make:
Logic Mnchines: Reasoning, Syllogism
Machine, Intelligence Testing. GamepI nyi ng Mach ines: Nim, Ti t-tat-toe •
Ari thmetic Machines: Adding, S u btracting, Multiplying, Divid in g,
Carrying, etc • .!d;:yp.1Q.grapldc ~bchines:
Secret Coder and Decoder, Combination
Locks. Puzzle Machines: The Space
Ship Airlock, The Fox lIen Corn and
Hired Mnn, Douglas Macdonald's Will,
The Uranium Ship and the Space Pirates.
--------Mail this Request-------or a Copy of It
Berkeley Enterprises, Inc,
~13 Avenue of the Americas, R143
Nell York 11, N. Y.
Please send me Geniac Kit llo. 1 and
Manual. Price, $17.95 (add 80¢ for
shipment in U. S. west of Mississippi,
$1.80 for shipment outside U. S. )
1 enclose
in full
payment. (If in good condi tion, it
is returnable in seven days for full
refund.) My name and address are
attached.

Qiji(}/C lOOIC
at your data with a

IpOTTERI DIGITAL RECORDER
You can "look into" your digital data-processing equipment conveniently and with
complete assurance of accuracy, using a Potter Model 960 Digital i Recorder. Permanent,
easily-stored records, capable of quick visual interpretation, -may be made of electrical
phenomena without interrupting the operation of equipment being monitored.
SPECIFICATIONS:
NO. OF CHANNELS Up to 32 on standard models; more on special order
RECORDING SPEEDS 150 measurements per second-continuous; up to 10 per secondintermittent
CHANNEL SENSITIVITY -25 voits-print; 0 volts-no print
RECORDING MEDIUM "Teledeltos" sensitized paper; standard 400-foot roll provides up
, to 52,000 measurements
PAPER TRAVEL Adjustable from front panel, 21/2" to 20" per second; continuous or
intermittent
Potter Digital Recorders are being used daily to monitor computer operation,
digital data-processing equipment and other applications. With minor modifications,
standard recorders can be provided with different input voltages, higher recording speeds
and more channels. Write today describing your "quick look" data recording -requirements.
Let our engineers suggest an appropriate solution; no obligation, of course.

POTTER INSTRUMENT COMPANY, INC.
115 Cutter Mill Road, Great Neck, N. Y.
- 41 -

Visit Us At The I.R.E. Show. Booth 346.
Computer Avenue

COMPUTERS A,ND AUT'OMATION - Back 'Copies
ARTICLES, PAPERS, ETC. January, 1955: Statistics
and Automatic Computers -- Gordon Spenser
Eastern Joint Computer Conference, Philadelphia,
Dec. 8-10,1954 -- Milton Stoller
The Digital Differential AnalY2er -- George F.
Forbes
A Small High-Speed Magnetic Drum - M. K. Taylor
An Inside-Out Magnetic Drum -- Neil Macdonald
February: Problems for Students of Computers
John W. Carr, III
Recogni2ing Spoken Sounds by Means of a Computer -- Andrew D. Booth
The Significance of the New Computer NOR C
W. J. Eckert
The Finan-8eer -- E. L. Locke
Approaching Automation in a Casualty Insurance
Company -- CarlO. Orkild
March: Question -- Isaac Asimov
--COmputers and Weather Prediction -- Br u c e
Gilchrist
Random Numbers and Their Generation - Gord 0 n
Spenser
Problems lnvol ved in the Application of Electronic Digital Computers to Business Operations -- John M. Breen
Computers to Make Administrative Decisions? -Hans Schroeder
April: Thinking Machines and Human Personality-Elliot L. Gruenberg
Marginal Checking -- An Aid in Preventive Maintenance of Computers -- J. ~~lvin Jones
~:
Reliability in Electronic Data Process 0 r s
-- William B. Elmore
Numerical Representation in Fix e d - Poi n t
Computers -- Beatrice H. Worsley
Automation -- A Report to the UAW-ClO Economic
and Collec.ti ve Bargaining Conference
The Skills of the American Labor Force -- James
P. Mi tchell
Automation Puts Industry on Eve of Fantas tic
Robot Era -- A. H. Raskin
The Monkey Wrench -- Gordon R. Dickson
June: THE COMPUTER DIRECTORY, 1955 (164 pages):
--Part 1: Who's Who in the Computer Field
Part 2: Roster of Organi2ations in the Co mputer Field
Part 3: The Computer Field: Prod uc ts and
Services for Sale.
July: Mathematics, the Schools, and the Oracle-Alston S. Householder
The Application of Automatic Computing Equipment to Savings Bank Operations - R. Hu n t
BrOliTn
The Book Reviewer -- Rose Orente
Linear Progrannning and Computers, Par t I -Cpandler Davis
August: The Automation of Bank Check Processing-R. Hunt Brown
Linear Programming and Computers, Par t II
Chandler Davis
Justifying the Use of an Automatic Computer
Ned Chapin
Charting on Automatic Data Process ing Sys tern s
-- Harry Eisenpress, James L. McPherson, and
Julius Shiskin
A Rotating Reading Head for Magnetic Tape an d
Wire -- National Bureau of Standards

Some Curiosities of Binary Arithmetic Usef u 1
in Testing Binary Computers -- Andrew D. Booth
September: A Big Inventory Problem and the IBM
702 - Neil Macdonald
Pq,lllications for Business on Automatic Computers:
A Basic Listing - Ned Chapin
Franchise -- Isaac Asimov
Automatic Coding for Digltal, Computers - G. M.
Hopper
Automatic Programming: The A-2 Compiler System
-- Part 1
October: The Brain and Learned Behavior -- Dr.
Harry F. Harlow
Automatic Programming: The A-2 Compiler Syste m
-- Part 2
Who Are Manning the New Computers? -- John M.
Breen
November: Automatic Answering of Inquiries - L.
E. Griffith
Found -- A "Lo,st" Moon -- Dr. Paul Herget
Mister Andrew Lloyd -- R. W. Wallace
December: Digital Computers in Eastern Europe
Alston S. Householder
Automatic Airways -- Henry T. Sinunons
Roster of Organi2ations in the Computer Field
(cumulative)
January, 1956: Machines and Religion -- E 11 i 0 t
Gruenberg.
Automatic Coding Techniques for Business Oat a
Processing -- Directions of Developmen t
Charles W. Adams, Bruce Moncreiff
What is a Computer? -- Neil D. Macdonald
REFERENCE INFORMATION (in various issues):
Roster of Organi2ations in the Computer Field /
Roster of Automatic Computing Services / Roster of Magadnes Related to Computers and
Automation / Automatic Computers: List /.Automatic Computers: Estimated Commercial Population / Automatic Computing Machinery: List 0 f
Types / Components of Automati c Computing
Machinery: List of Types / Products and Se rvices in the Computer Field / Who's Who in
the Computer Field / Automation: Lis t 0 f
Outstanding Examples / Boo k s . and 0 the r
Publications / Glossary / Patents I Titles
and Abstracts of Papers
BACK COPIES: Price, if available, $1.25 eac h,
except June, 1955, $4 .00 ~ Vol. 1, no. 1, Sept.
1951, to vol. 1, no. 3, July, 1952: 0 u t 0 f
print. Vol. 1, no. 4, Oc t. 1952: in pri n t.
Vol. 2, no. 1, Jan. 1953, to vol. 2, no. 9,
Dec. 1953: in print ex c e p t Mar c h, no. 2,
May, no. 4_. and July, no. 5. Vol. 3, no. 1,
Jan. 1954, to vol. 3, no. 10, Dec. 1954:
in
print. Vol. 4, 1955, no. 1 to 12, in print.
A subscription (see rates on page 4) may be specified to begin with the current month's or the
preceding month's issue.
WRITE TO:

Berkeley Enterprises, Inc.
Publisher of COMPUTERS AND AUTOMATION
513 Avenue 'of the Americas
New York 11, N. Y.
- 42 -

Simulation
Engineering

2 ENGINEERS
ME or EE

; DIGITAL
COMPUTER DEVELOPMENT
l

Excellent opportunity to join an expanding, stable company, with an
outstanding position in the precision
electronic' control industry.
Opening now available for an engineer able to assume responsibility
for the d eve lop men t of complete
systems for fire control and guidance, or major portions of such systems. Work will include research and
development in the field of complex
analog or digital computers.
A degree in' electrical or mechanical
engineering is required or the equivalent in experience.

To conduct system studies on RO<4{ETS,
MISSILES, and HIGH SPEED AIRCRAFT using the latest high speed
digital computers and the most up-todate analog computers.
Attractive opportunities exist for both
experienced personnel and recent graduates in the following fields:

MISSILE &
WEAPONS CONTROL
SYSTEMS STUDIES

• HIGH SPEED
DIGITAL LOGIC

Enjoy full use of your skill and
imagina tion and friendly professional
give-and-take with top men in the
field of electro-mechanical precision
equipment.
You will perform studies related to
airborne weapons control and guidance systems with the object of determining requirements, feasibility,
performance and specifications of
computers and overall systems.
A degree in physics, ME or EE - or
the equivalent in experience is
required. Must be able to handle
problems in such diversified fields
as digital computers, digital data
transmission systems, logic counting
and conversion circuits, high-precision gyro and gimbal structures,
dynamic behavior of missiles, alignment of inertial platforms for guidance systems and fight evaluation of
guidance systems and instrumentation. Send resume in confidence to

• AUTOMATIC
CONTROL THEORY
• ANALOG COMPUTATION
TECHNIQUES
Please Contact
Professional Employment

Technical Personnel Dept. 2.500

ARMA

1VI~.it

Division American Bosch Arma Corp.
Roosevelt Field, Garden City

BALTIMORE

Long Island, N. Y.

- 43 -

3.

»
MARYLAND

ADVERTISING IN ~~COMP'UTERS AND A'UTiOMATION"
~ffemorandum from Berkel ey Enterprises, Inc.
Publisher of CO~lPUTERS ,\ND ATJTO\fATION
36 West 11 St., New York 11, N. Y.

1. What is "COMPUTERS AND AUTOMATION"? It is
a montly magazine containing articles, papers,
and reference information related 'to computing
machinery, robots, automatic control, cybernetics, automation, etc. One important piece of
reference information published is the "Ros t e r
of Organizations in the Field of Computers and
Automation". The basic subscription rate is
$5.50 a year in the Uni ted States. Sin g 1 e
copies are $1.25, except June, 1955, "The Computer Directory" (164 pages, $4.00). For the
titles of articles and papers in recent issues
of the magazine, see the "Back Copies" page in
this issue.
2. What is the circulation? The circulation
includes 2000 subscribers (as of Dec. 10): over
300 purchasers of individual back copies; and
an estimated 2500 nonsubscribing readers. The
logical readers of COMPUTERS AND AUTOMATION
are people concerned wi th the field of computers
and automation. These include a great number
of people Ivho tvill make recommendations to their
organizations about purchasing computing machinery, similar machinery, and components,
and whose decisions may involve very substantial figures. The print order for the Jan.
issue was 2500 copies. The overrun is largely
held for eventual sale as back copies, and in
the case of several issues the over ru n has
been exhausted through such sale.

copy that may be put under the ph oto 0 f f s e t
camera without further preparation. Unscreened
photographic prints and any other copy requirfug
addi tional preparation for photooffset shou 1 d
be furnished separately; it will be prep ared,
finished, and charged to the advertise r a t
small addi tional costs. In the case of printed
inserts, "a sufficient quantity for the iss u e
should be shipped to our printer, addres s on
request.
Display advertising is sold in units of a full
page (ad size -(" x 10", basic rate, $190) twothirds page (basic rate, $145), and half pag e
(basic rate, $97); back cover, $370; in sid e
front or back cover, $230. Extra for color red
. (full pages only and only in certain posi tions),
35%. Two-page printed insert (one sheet) , $32U;
four-page printed insert (two sheets), $590.
Classified advertising is sold by the t\' 0 r d
(60 cents a word) wi th a minimum of 20 words.

3. What type of advertising does COMP UTE RS
AND AUTOMATION take? The purpose of the magazine is to be factual and to the point. For
this purpose the kind of advertising Ivanted is
the kind that answers questions factually. We
recommend for the audience that we reach, that
advertising be factual, useful, interesting,
understandable, and new from issue to issue.
We reserve the right not to accept advertisfug
that does not meet our standards.
4. What are the specifications and cost 0 f
advertising? COMPUTERS AND AUTOMATION is published on pages 8~" x 11" (ad size, 7" x 10")
and produced by photooffset, except that printed sheet advertising may be inserted and bound
in with the magazine in most cases. The closing date for any issue is approximately th e
10th of the month preceding. If possible, the
company advertising should produce final copy.
For photooffset, the copy should be exact 1 y
as desired, actual size, and assembled, and
may include typing, writing, line drawing,
printing, screened half tones, and any othe r

- 44 -

5. Who are our advertisers? Our advertisers
in recent issues have included the follow in g
companies, among others:
Ampex Corp.
Arnold Engineering Co.
The Austin Co.
Automatic Electric Co.
Bendix Aviation Corp.
Cambridge Thermionic Corp.
Epsco, Inc.
Ferranti Electric Co.
Ferroxcube Corp. of America
General Electric Co.
Hughes Research and Development Lab.
International Business Machines Corp.
Lockheed Aircraft Corp.
Logistics Research, Inc.
The Glenn L. Martin Co.
Monrobot Corp.
Norden-Ketay Corp.
Northrop Aircraft, Inc.
George A. Philbrick Researches, Inc.
Potter Instrument Co.
Raytheon Mfg. Co.
Reeves Instrument Co.
Remington Rand, Inc.
Republic Aviation Corp.
Sprague Electric Co.
Sylvania Electric Products, Inc.

ROBOT
SHOW
STOPPERS
From time to time you may nee d t 0
help organize a display in a business
show including some device that you
hope will "STOP" every perso n attending the show and make him notice
your display - a device which may
be called a "SHOW srOPPER".
In addition to publishing the magazine "COMPUTERS AND AUTOMATION", we
have for five years been developing
and constructing "ROBOT SHOW STO PPERS", small robot mach i n est hat
res pond to the ir environmen t a n d
behave by themselves.
TtfO of them
are:

INGENIOUS
PACKAGING

RELAY MOE: A machine that will play
the game Ti t-Tat-Toe with a human
being, and either win or draw all
the time, or (depending 0 nth e
setting of a switch) will sometimes lose, so as to make the game
more interesting for the hum an
being;
FRANKEN: A mechanical rat that will
explore a maze, find "food", and
learn the path through; the maze
may be set up by any pers on in the
audience, using little partitions
in any way that he wants to.
Bes ides these we have other sma 11
robots finished or under development.
Some of these machines have been on
the front covers of the magazine s
"Scientific American" and "R a d i 0
Elec tronics". These machines may be
rented for shows under certain conditions; also, modifications of the
small robots to fit a particul CJ r
purpose are often possible, such as
use of particular components, d i splay of particular slogans, etc.
---------~----~--------~--------~

To: Berkeley Enterprises Inc.,
36 West 11 St., R143,
New York 11, N. Y.

Please send us more informa ti 0 n
about your ROBOT SHOW STOPPERS. The
advertising application we have i n

Further applications of
electrom6chanical techniques in these fields
are creating new openings in the
Systems Division of Hughes Research
and Development Laboratories.

The most
advanced
. developments
in electronics
are being
made in
the sphere of

Engineers who have demonstrated ingenuity and inventive
ability will find interest in areas of work that call for
devising reliable, maintainable, manufacturable designs
for precision equipment developed at Hughes Research and
Development Laboratories.

airborne radar
and related
ground control

The design of this equipment, manufactured at Hughes,
involves ·mechanical, electromechanical, electronic, microwave and computing problems. Design also requires the
use of such advanced techniques as sub miniaturization,
unitized "plug-in" construction, with emphasis on design
for volume production. Knowledge of electronic components,
materials, finishes and military specifications is useful.

systems because
of military
emphasis.

SCIENTIFIC STAFF RELATIONS

mind is: ________________________
From: (Organization)

HUGHES

(Address)

RESEARCH AND DEVELOPMENT LABORATORIES

(Filled in by: Name, Title, Date)

Culver City, Los Angeles County, California

- 45 -

ADVERTISING INDEX
The purpose of COMPUTERS AND AUTOMATION is to be
fac tual, useful, and unders tandable. For this purpose, the kind of advertising we desire to publish
is the kind that answers questions, such as: What
are your products? What are your services: And
for each product, What is it called? What does it
do? Ilot. . t. . ell does it t. . ork? What are its m a i n
spec if ications '?

Ferranti Electric Inc., ~O Rockefeller Plaza, New
York 2~, N. Y. / High Speed Tape Reader / page
39 / CA No. 85
.
Ferroxcube Corp., East Bridge St.,/Saugerties, N. ~
/ Magnetic Core Materials / page 36 / CA No. 86
:Iughes Research and Development Laboratories, Culver City, Calif. / Help Wanted / page 45 / CA
No. 87
Lockheed Aircraft Corp., California Div., Burbank,
Calif. / Missile Systems Mathematics / page 47/
CA No. 88
The Glenn L. Martin Company, Baltimore 3, Md. /
Simulation Engineering / page 43 / CA No. 89
Potter Instrument Co., 115 Cutter Mill Rd., Great
Neck, N. Y. / Digital Recorder / page 41 / CA
No. 90
Remington Rand, Inc., 315 4th Ave., New York 10,
N. Y~ / Univac / page 2 / CA No. 91
Sylvania Electric Products, Inc., 175 Great Arrow
Ave •• Buffalo 7, N.
niin Triode / page 48/
CA No. 92

Following is the index and a summary of advertisements. Each i tern contains: Name and address 0 f
the advertiser / subj ec t of the advertisement /
page number where it appears / CA number in case
of inquiry (see note below)Arma Division, American Bosch Arma Corp., Roosevel t Field, Garden Ci ty, L. I., Net~ York / Engineering Opportunities / Page 43 / CA No. 79
Armour Research Foundation of Illinois, Institute
of Technology, Technology Center, 'Chicago 16,
Ill. / Help Wanted / page 38 / CA No. 80
The Arnold Engineering Co., Marengo, Ill. / Tapet'l/Ound Bobbin Cores / page 33 / CA No. 81
Automatic Electric Company, 1033 W. Van Buren St~
Chicago, Ill. / Circuits / page 5 / CA No. 82
Berkeley Enterprises, Inc., 513 Ave. of the Americas, Net~ York 10, N. Y. / Publications, Gen:i£c,
Robot Show Stoppers / pages 39, 41, 45 / CA
No. 83
ComplJ.1.~~rs and Automation, 513 Ave. of the Amer~as.
Net~ York 10, N. Y. / Gateway to Science, Back
Copies, Advertising / pages 32, 42, 44 / CA
No. 84

r. /

READER'S IN'QUIRY
If you wish more information about any products
or serVIces mentioned in one or more of these
advertisements, you may circle the appropriate
CA Nos. on the Reader's Inquiry Form below and
send that form to us (we pay postage; see the
instructions). We shall then forward your inquiries, and you will hear from the advertisers
direct. If you do not wish to tear the magazine,
just drop us a line on a postcard.

*--------------------------------------------*-------------------------------------------*
READER'S INQUIRY FORM
Paste label on envelope:~
- - --- - - - . , - - -- --..,.~-

---

~

"A-.."
0

'"......

ell

m

~

::II;

m
rm

-<

m

w

0-

zCD

~

...
-< ...
.. :r......
..
...
:z:
CD

CIt

~

0

~

~

CIt

z
....
m
~

-a

~

Vi

m
,!II

QI
~

CD

~
to

CD

.."
QI

a.:

- - ... - -

-

-- - - -

-

--

z tD
0
."

0

~
G)
1ft
CIt

~

...

C

C/')

Z

.zp;

INQUIRY

READER,'S

FORM

Name (please print) ........................................................................................................................... .

m

3:: C/')

Your Address? ..................................................................................................................................... .

C/')

Your Organization? .......................................................................................................................... .

c:

;::c

;JIll

m

Its Address? .................................................... _.................................................................................. .

"-<

Your Title ? ..................................................... ,.................................................................................... .

~

-<

:;;
3::

>

;::
m
0

r-

r
z~

Please send me additional information on the following subiects for which I have
circled the CA number:

J

O:J
m

1
6
11
16
21

r-

to

z "<
!l

2
7
12
17
22

3
8
13
18
23

4
9
14
19
24

5
10
15
20
25

26
31
36
41
46

2J
32
37
42
47

28
33
38
43
48

29
34
39
44
49

30
35
40
45
50

51
56
61
66
71

52
57
62
67
72

53
58
63
68
73

54
59
64
69
74

55 76 n 78 79 80 101 102 103 104 105
60 81 82 83 84 85 106 107 108 109 110
65 86 87 88 89 90 111 112 113 114 115
70 91 92 93 94 95 116 117 118119 120
75 96 97 98 99 100 121 122 123 124 125

126
131
136
141
146

127 128
132133
137138
142143
147 148

129
134
139
144
1-4Y

130
135
140
US
150

ZVl ."

CD
CD

mO

~p

w

ITI

;:u ::!!

-<~ 3:
0-

:<

::0"'0

?,'

111111111111111

~_.

=i

Z
z· 0
• !Ie
:
Source Exif Data: 
File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.3
Linearized                      : No
XMP Toolkit                     : Adobe XMP Core 4.2.1-c041 52.342996, 2008/05/07-21:37:19
Create Date                     : 2015:08:22 08:04:34-08:00
Modify Date                     : 2015:08:22 07:22:58-07:00
Metadata Date                   : 2015:08:22 07:22:58-07:00
Producer                        : Adobe Acrobat 9.0 Paper Capture Plug-in
Format                          : application/pdf
Document ID                     : uuid:3b2d69f3-df64-e742-82aa-459decc006dc
Instance ID                     : uuid:c3818d8d-999b-5a46-b995-407bd5305a2b
Page Layout                     : SinglePage
Page Mode                       : UseNone
Page Count                      : 48

EXIF Metadata provided by EXIF.tools