Books offered by JF Ptak Science Books

Foundations for the First All-transistor Computer, 1950/1951

Among the Earliest Diagrams for a transistorized Computer  Ca. 1950/1

 

J.H. Felker.  Catalog of Digital Computer Designs.  [No place of publication and no date.]  I suspect that this is a Bell Laboratories (Whippany, New Jersey), publication which was printed in 1950/1951. [On dating this item:  I’m thinking that this paper was released before his other two papers in late 1951 as they are not referenced in the notes section.  Also there is no reference in the paper to an actual delivery date of the transistors, which Bell and Western Electric announced would be available “several months after” the first transistor conference in 1951.  Also the transistor pulse amplifier which Felker requires in his conclusions section seems to have been not available until 1953.]

 11x8 ½ inches.  18 pages of text, 18 leaves of diagrams and schematics.  All text and drawings are printed on one side of the page, only.  This seems to be offset-printed.  It also seems to be made for restricted circulation.  Very Good condition.  $3000 

 

 This is Felker’s OUTSTANDING orientation on constructing a transistorized computer.* He states in the second paragraph “this computer design philosophy was followed in the design of the National Bureau of Standards Computer SEAC. It is believed that the approach that will result in a vacuum-tubeless computer at the earliest date is to follow the SEAC example in so far as the use of germanium diode logic circuits is concerned, but replacing the vacuum tube amplifiers with transistor amplifiers.”  He states further:  “Since the transistor itself has voltage and current relationships quite similar to a germanium diode it is expected that the diodes in a transistor computer will operate in a more natural environment…and will exhibit…longer life and more reliable operation.”

 As a matter of fact all of the block diagrams in the illustration section are for “an all-semiconductor computer”.  Essentially these are among the earliest printed diagrams for a transistorized computer—and Felker would be the leader of the Bell team that constructed the world’s first all-transistor computer, the TRADIC, in 1955.

Felker was the leader of the Bell team that constructed the TRADIC, which was the world’s first fully transistorized computer.  It used 800 transistors and 10,000 germanium triodes and dispensed with vacuum tubes—it was much smaller (at three cubic feet) than any of its contemporaries, and was at least as fast as the fastest computer of the day.  It was also far more reliable, and far more dependable.

Felker concludes his paper thus:  “A fairly complete catalog of digital computer components has been given in block diagram form with estimates of the number of parts they require…  The principal conclusion to be drawn is that an all semiconductor computer can be built with diodes and transistors as soon as a transistor pulse amplifier with a high degree of reliability is available…”

The paper is structured in the following way, by sections:  (Synopsis and introduction); Basic Building Blocks; Switches; Handling of Negative Numbers; Adder; Accumulator; Multiplier; Divider; Translator; Binary Counter; Shift Registers; conclusions.

[The image to the right is folding, and thus partly obscured.]

The list of block diagrams, as follows:  Or-circuit; and-Circuit; Inhibitor Circuit; Amplifier with Pulse retiming; Storage Cell; a Simple Switch; an Eight Position Switch; Handling of Negative Numbers; Adder; Accumulator; Multiplier; Operation of Multiplier; Binary Division; Simplified Division Process; Divider; Translator; Binary counter; Shift register.

NOTES:

*For Felker's TRADIC Phase I computer project. See James R. Harris, "The Earliest Solid-State Digital Computers," IEEE Annals of the History of Computing, vol. 21, no. 4, pp. 49-54, Oct.-Dec. 1999, doi:10.1109/85.80153  AND  P.K. Giloth and R.C. Townley, TRADIC—Flyable Research Model Program, Summary Engineering Report, U.S. Air Force, Air Materiel Command, Dec.1 1958.

J.H. Felker, "Performance of the TRADIC Transistor Digital Computer," Proc. Eastern Joint Computer Conf., 1954, pp. 46-49. J.R. Harris, "TRADIC: The First Phase," Bell Laboratories Record, Vol. 36, Sept. 1958, pp. 330-334.

Felker.  Application of Transistors in a High-Speed Computer. Bell Labs Transistor Symposium, November 1951. (This is the event where MR. Felker was said to have amazed the other high-power attendees with his talk on transistor applications to the digital computer. 

Felker.  “The transistor as a digital computer component.”  AFIPS Joint Computer Conferences  Papers and discussions presented at the Dec. 10-12, 1951, joint AIEE-IRE computer conference: Review of electronic digital computers .

This is a picture of Mr. Felker, (kneeling), before the TRADIC. 

The Butterfly Effect, Chaos, and the Weather: Edward Lorenz, 1967

JF Ptak Science Books LLC  (Book store website)
History of Ideas Blog

Edward Lorenz.  The nature and theory of the general circulation of the atmosphere. 1967 [Part 1] [Part 2] [Part 3]  World Meteorological Organization, No. 218, TP 115, 161 pp.   $500.00

This is Lorenz's (1917-2008) masterpiece on the dynamics of the atmosphere, which is perhaps the greatest work on the weather since it was first scientifically treated by George Hadley in the 1730's.  What Lorenz presents here is his theoretical treatment of the non-linear action of the circulation of the atmosphere, an accidental discovery he made* in 1963 ("Deterministic Nonperiodic Flow in Journal of the Atmospheric Sciences" see below)  which resulted in the establishment of the foundations of modern chaos theory.

This is a decent, working copy of his book, which seems to be quite scarce--it doesn't show up very often in collections in universities, and is basically absent from auction records and bookseller catalogs.  It is also only the second copy that I've had--both have been ex-library copies.  This copy is generally crisp and bright, and has only a few scant library marks.  This has an interesting library association, coming as it does from the Institute for Defense Analysis (IDA) library in Arlington.  Given the early work by Lorenz for the US Army Air Force and his subsequent work that was of high interest to the military, the association is appropriate. 

*From a wiki article, a nice description of the accidental nature of Lorenz's discovery:

"An early pioneer of the theory was Edward Lorenz whose interest in chaos came about accidentally through his work on weather prediction in 1961.[14] Lorenz was using a simple digital computer, a Royal McBee LGP-30, to run his weather simulation. He wanted to see a sequence of data again and to save time he started the simulation in the middle of its course. He was able to do this by entering a printout of the data corresponding to conditions in the middle of his simulation which he had calculated last time.

"To his surprise the weather that the machine began to predict was completely different from the weather calculated before. Lorenz tracked this down to the computer printout. The computer worked with 6-digit precision, but the printout rounded variables off to a 3-digit number, so a value like 0.506127 was printed as 0.506. This difference is tiny and the consensus at the time would have been that it should have had practically no effect. However Lorenz had discovered that small changes in initial conditions produced large changes in the long-term outcome.[15] Lorenz's discovery, which gave its name to Lorenz attractors, proved that meteorology could not reasonably predict weather beyond a weekly period (at most)."

Significant Works by Lorenz  (a full listing from MIT is found here.) 

1955 Available potential energy and the maintenance of the general circulation. Tellus. Vol.7
1963 Deterministic nonperiodic flow. Journal of Atmospheric Sciences. Vol.20 : 130—141 link [10].
1967 The nature and theory of the general circulation of atmosphere. World Meteorological Organization. No.218
1969 Three approaches to atmospheric predictability. American Meteorological Society. Vol.50
1976 Nondeterministic theories of climate change. Quaternary Research. Vol.6
1990 Can chaos and intransitivity lead to interannual variability? Tellus. Vol.42A
2005 Designing Chaotic Models. Journal of the Atmospheric Sciences: Vol. 62, No. 5, pp. 1574–1587.

Video of Lorenz Speaking

Predecessors to Lorenz in Chaos Theory and Dynamical Systems

Jules Henri Poincaré (1890) "Sur le problème des trois corps et les équations de la dynamique. Divergence des séries de M. Lindstedt," Acta Mathematica, vol. 13, pages 1–270.

Hadamard, Jacques (1898). "Les surfaces à courbures opposées et leurs lignes géodesiques". Journal de Mathématiques Pures et Appliquées 4: pp. 27–73.

George D. Birkhoff, Dynamical Systems, vol. 9 of the American Mathematical Society Colloquium Publications (Providence, Rhode Island: American Mathematical Society, 1927)

Kolmogorov, A. N. (1941a). “Local structure of turbulence in an incompressible fluid for very large Reynolds numbers,” Doklady Akademii Nauk SSSR, vol. 30, no. 4, pages 301–305. Reprinted in: Proceedings of the Royal Society of London: Mathematical and Physical Sciences (Series A), vol. 434, pages 9–13 (1991).

Significant Papers by Loenrz in pdf

The nature and theory of the general circulation of the atmosphere. [Part 1] [Part 2] [Part 3]  World Meteorological Organization, No. 218, TP 115, 161 pp.

and

Edward N. Lorenz (1963). "Deterministic Nonperiodic Flow". Journal of the Atmospheric Sciences 20: 130–141.http://ams.allenpress.com/archive/1520-0469/20/2/pdf/i1520-0469-20-2-130.pdf.
Edward N. Lorenz (1969). "Atmospheric predictability as revealed by naturally occurring analogues". Journal of the Atmospheric Sciences 26: 636–646. doi:10.1175/1520-0469(1969)26<636:APARBN>2.0.CO;2. http://ams.allenpress.com/archive/1520-0469/26/4/pdf/i1520-0469-26-4-636.pdf.
Edward N. Lorenz (1969). "Three approaches to atmospheric predictability". Bulletin of the American Meteorological Society 50: 345–349. http://eapsweb.mit.edu/research/Lorenz/Three_approaches_1969.pdf.

Notices on Lorenz

Tim Palmer (2008). "Edward Norton Lorenz". Physics Today 61 (9): 81–82. doi:10.1063/1.2982132.
"Lorenz Receives 1991 Kyoto Prize". MIT News Office. 1991. http://web.mit.edu/newsoffice/tt/1991/24996/24998.html.

Awards

1969 Carl Gustaf Rossby Research Medal, American Meteorological Society.
1973 Symons Memorial Gold Medal, Royal Meteorological Society.
1975 Fellow, National Academy of Sciences (U.S.A.).
1981 Member, Norwegian Academy of Science and Letters.
1983 Crafoord Prize, Royal Swedish Academy of Sciences.
1984 Honorary Member, Royal Meteorological Society.
1991 Kyoto Prize for ‘… his boldest scientific achievement in discovering "deterministic chaos" .’.

Two Classic Works in Cybernetics--Wiener, 1948 and Ashby, 1956

I'm happy to offer a pair of significant books in cybernetics:

Wiener, Norbert.  Cybernetics, or Control and Communication in the Animal and the Machine. 

Published by the Technology Press of MIT, 1948.  First edition.  This is a beautiful copy in a VERY GOOD (somewhat chipped) dustjacket.  Wiener of course was the sensationally gifted mathematician at MIT where, among other things, coined the word "cybernetics".  ("Wiener was a pioneer in the study of stochastic and noise processes, contributing work relevant to electronic engineering, electronic communication, and control systems. Wiener also founded cybernetics, a field that formalizes the notion of feedback and has implications for engineering, systems control, computer science, biology, philosophy, and the organization of society.")

OFFERED WITH:

Ashby,  W. Ross:  An Introduction to Cybernetics;  London:  Chapman and Hall,  1956.  1st edition.   
295pp  8vo.  Cloth.  Very good condition.   

The pair:  $850.

W. Ross Ashby (London, 1903-1972)  was an English psychiatrist and a pioneer in cybernetics and the study of complex systems.

REVIEW COPY. Lovely copy in a near-fine, crisp, colorful  dustjacket. Previous owner's inscription. Review Copy slip repaired. Scarce--this is the only review copy of this book we have seen.

Ross Ashby was one of the original members of the Ratio Club, a small informal dining club of young psychologists, physiologists, mathematicians and engineers who met to discuss issues in cybernetics. The club was founded in 1949 by the neurologist John Bates and continued to meet until 1958.

Earlier, in 1946, Alan Turing wrote a letter to Ashby suggesting he use Turing's Automatic Computing Engine (ACE) for his experiments instead of building a special machine. In 1948 Ashby made the Homeostat.

The High-Popular Appraisal of Computers, 1949-1950

 I offer the four wonderful and somewhat significant appearances of Claude Shannon, Edmund Berkeley, Warren Weaver and Harry M. Davis as they write about cutting-edge technology in the computer world for 1949/1950.  It is very interesting to see them in situ, giving the coverage of the latest breaking news in the compsci world a real taste of local color.
The four, all in original (colorful!) wrappers, in Good condition: $1000

Berkeley,  Edmund C.:  "Simple Simon";  New York:  Scientific American,     Vol 183, No. 5, November 1950  Pp 40-44    Original printed wrappers.  Near fine condition.          This is the entire issue of SA for November 1950 and features Columbia University's "electrical brain" in great splendor (and in color) on the fron cover. Striking.   "Last spring the world's smallest true mechanical brain was unveiled at Columbia University.  Its function:  to explain in a rudimentary way the principles and the operation of its forbidding and inaccessible relatives."

Offered With:

Warren Weaver.  "The Mathematical Theory of Communication".  New York, published in Scientific American.  Vol 181, No. 1, July 1949.  Pp 11-15.  Good condition, nice and bright copy.  This is the entire issue for July 1949. 

Offered With:

Harry M. Davis. "Mathematical Machines"; New York, Scientific American, Vol 180 No. 4; April 1949.  Occupies pp 28-40.  We offer the entire issue for April 1949.  "The electrical and mechanical devices that have assumed certain functions of the human brain are rapidly becoming a big business.  A review of the present state of affairs, with emphasis on the digital computer."

And Offered With:

Claude E. Shannon.  "A Chess Playing Machine", New York, Scientific American, Vol 182, No. 2.  Occupies pp 48-51.  "The modern electronic computer can be set up in such a way as to play a fairly strong and apparently imaginative game of chess against a human opponent.  This raises the question of whether it can "think". 

Excellent Review of Computers, 1953

Booth, Andrew and Kathleen Booth,  Automatic Digital Calculators.
London:  Butterworths,  1953.  1st edition.  Green buckram with gold lettering.  Very fine condition.  Very fine dust jacket.  Beautiful copy of the first edition in an excellent dj.  $350

This is truly a superior effort at a survey of the state of computation via digital computer for the Post WWII-1953 era.  The 17 chapters are geared mainly as an instructional--an advanced "how to", if you will, with plenty of diagrams and illustrations.  After a few historical chapters, we have:  the overall design of a computing system; the control; the arithmetic unit; miscellaneous operations; input-output; gates; single digit storage; miscellaneous components, storage devices.  From this point on, from p136-196, the book deals primarily with programming issues:  definitions of a code and discussion of its form and controls (!), pp 136-151; the techniques of coding; the use of subroutines in coding; programme design; some applications of computing machinery. 

I should also point out that the very tight, compacted bibliography section (occupying pp 217-226) is particularly useful, for on pp 219-22 is found a sub-bibliography arranged by computer name, identifying 37 different computers (the Zuse computers for example would comprise just "one" computer). 

The Pioneering "Numerical Analysis" by D.R. Hartree

Hartee, Douglas R.  Numerical Analysis.  Oxford, 1952.  287pp, octavo.  Beautiful copy with a very good issue of its scarce dustjacket. (The Oxford just jackets are notorious for fading along the spine--this one has too, but not by very much.)  Lovely copy of this book not known for being found in lovely condition.  $350

The Dictionary of Scientific Biography has this to say about numerical analysis and the great computing pioneer Hartree:

"Hartree's chief contribution to science was his development of powerful methods of numerical mathematical analysis, which made it possible for him to apply successfully the so-called self-consistent field method to the calculation of atomic wave functions of polyelectronic atoms, that is, those which in the neutral condition have more than one electron surrounding the nucleus. These calculations involved the numerical solution of the partial differential equations of quantum mechanics for many-body systems subject to the usual boundary conditions" (Volume 6, p. 147).

Chapter XII, the last chapter in the book, is titled "Organizations of Calculations for an Automatic Machine" and covers pages 255-273. It is an excellent summary of the state of affairs for computing though it contains just one footnote in the chapter.   In general, though the four-page bibliography is very useful. 

The Hagelin Cryptograph Machines: Secure Communications, 1942

[Offered in conjunction with our website, JF Ptak Science Books LLC]

[Hagelin]  The Hagelin Cryptographers, an Analysis.
Stamped  CONFIDENTIAL;  printed in New York:  Ericsson Telephone,  1942.      19 leaves.     Very good condition.       
Mimeographed sheets, stapled.  11x8, 19pp. Offset, typed document. Stamped "Accessions Division, Nov 11, 1942, Library of Congress". With an accompanying cover letter with the rubber stamp of Ericsson Telephone, Sales Corp, NYC., and dated July 3, 1942.  RARE.  $1500

This is a general report on the origin, development and status of the Hagelin "cryptographers"-a word used here to describe the physical machines (rather than the people working on codes). Sections in the document include "Models Built at Express Demand of the French Authorities", "Evolution of Hand Cryptographer Type C-362", "Hagelin Cryptographer Models" (BC-38 and C-362), "Methods of Operation", "Superiority of Hagelin Cryptographers over Competing Makes", and others, including a final section "How to Sell Cryptographers". There is a mention of the "Enigma" machine on page 14, which is limited to mentioning that it is not sold outside of Germany. Although the Swiss firm founded by Boris Hagelin has manufactured, and continues to manufacture, many kinds of cipher machines, the words "Hagelin machine" will normally inspire thoughts of their unique lug and pin based machines.

"The basic principle of a Hagelin lug and pin machine is easy enough to describe. In the C-38, used by the U.S. Army as the M-209, six pinwheels, with 17, 19, 21, 23, 25, and 26 positions on them, can be set by the user with an arbitrary series of pins that are active. For every letter enciphered, all the pinwheels rotate one space. The combination of active and inactive pins is presented to a cage with 27 sliding bars. Each bar has two sliding lugs on it, which can be placed either in a position where it is inactive, or in a position corresponding to any of the pinwheels, so that it will slide the bar to the left, if the pin currently presented by that pinwheel is active. The number of lugs sticking out rotates the cipher alphabet against the plain text alphabet. The two alphabets used are just the regular alphabet, and the alphabet in reverse order, from Z back to A. This meant that encipherment was reciprocal, although the machine still had a switch to select encipherment or decipherment: this determined if the machine printed its output in five letter groups, or if it translated one letter, chosen by the user, to a space. The C-52, a postwar version of the Hagelin lug and pin machine, added an extra five sliding bars to the cage that, instead of moving the cipher alphabet, caused the stepping of the pinwheels to be irregular. The first pinwheel always moved, but the remaining five pinwheels only moved when their corresponding bars were slid to the left. The six pinwheels were labeled A, B, C, D, E, and F from left to right; bar 1 controlled pinwheel B, bar 2 pinwheel C, and so on. Also, on the C-52 the lugs could be moved from bar to bar, and the six pinwheels were chosen from a set with lengths 25, 26, 29, 31, 34, 37, 38, 41, 42, 43, 46, and 47. Using the pinwheels with lengths 34, 38, 42, 46, 25, and 26 allowed one to achieve compatibility with the C-36: provided one also turned off the irregular pinwheel stepping feature. The alphabet always started from its normal position, instead of the position last used, before being rotated by the projecting slide bars. This was perhaps the machine's main weakness, as it made attacks based on frequency counts of displacements possible, but it was perhaps unavoidable, since there was always a slight possibility of occasional mechanical errors. Particularly as the machines were often used on battlefields."   ABE403   Available 

Pre-histories of Great Ideas: Mandelbrot and the Fractal, 1967-1975

This is site operated in conjunction with JF Ptak Science Books LLC

Mandelbrot, Benoit.  "How Long is the Coastline of Great Britain?. " Offprint from Science Magazine, 1967.
A rarity!  10 x 8 inches, 8pp. Original printed wrappers.  Fine copy. $2000

This is the first of a few short essays of the idea of pre-historic appearances of important ideas. I use the word “pre-historic” in its literal sense—that is, these papers were published before the concepts that they discussed were recognized and published, later, in more complete form and of course to greater fame. One will certainly be Gregor Mendl’s paper (Versuche über Plflanzenhybriden ) published in the obscure Verhandlungen des naturforschenden Vereines in Brünn, (Brno), and to be resurrected at the hands of Gregory Bateson thirty-odd years later. Another example is Vannevar Bush’s “As We May Think”, published in The Atlantic Monthly in June 1945, and which was a definite intellectual precursor to the construction of the Internet. A third

Right now, though, I’d like to look at the pre-birth of fractals in Benoit Mandelbrot’s precursor (1967) paper (and in a weird, probably not-correct way, scooping himself) to his more famous effort of 1975, in which he coins the term and fully describes fractals, showing the measurement of the coastlines behave like a fractal over a range of measurement scales.

The seminal paper, How Long Is the Coast of Britain? Statistical Self-Similarity and Fractional Dimension, published in Science magazine in 1967, is an introduction to the concept of fractals (which Mandelbrot would name in later in1975), The kernel of it all is found in the Lewis Fry Richardson graphs and more so in the small illustration on page three of the paper—the peek into the future world of fractals, seeing a fractal before it was named so.

Richardson (a kind of Jevons, Marey-like American figure on the history of 20^th century science) did some very good thinking about the elusive nature of naming such a thing as the lengths of coastlines and other natural constructions: namely he determined that their measured length was dependent upon the scale of measurement. That is to say the measured length, L(G), is ruled the scale of G, approximating the function as L(G)=M(G)1-0 

Mandelbrot moves on to discuss the coastline in terms of it being a self-similar object, which is something that is nearly exactly similar to a part of itself, or the whole having the same shape of one or more than one of its parts, and regresses.

 

 

 

The Mandelbrot Fern (courtesy of http://en.wikipedia.org/wiki/Image:Fractal_fern_explained.png) displaying self-similar properties. 

This is the first of a few short essays of the idea of pre-historic appearances of important ideas. I use the word “pre-historic” in its literal sense—that is, these papers were published before the concepts that they discussed were recognized and published, later, in more complete form and of course to greater fame. One will certainly be Gregor Mendl’s paper (Versuche über Plflanzenhybriden ) published in the obscure Verhandlungen des naturforschenden Vereines in Brünn, (Brno), and to be resurrected at the hands of Gregory Bateson thirty-odd years later. Another example is Vannevar Bush’s “As We May Think”, published in The Atlantic Monthly in June 1945, and which was a definite intellectual precursor to the construction of the Internet. A third

Right now, though, I’d like to look at the pre-birth of fractals in Benoit Mandelbrot’s precursor (1967) paper (and in a weird, probably not-correct way, scooping himself) to his more famous effort of 1975, in which he coins the term and fully describes fractals, showing the measurement of the coastlines behave like a fractal over a range of measurement scales.

The seminal paper, How Long Is the Coast of Britain? Statistical Self-Similarity and Fractional Dimension, published in Science magazine in 1967, is an introduction to the concept of fractals (which Mandelbrot would name in later in1975), The kernel of it all is found in the Lewis Fry Richardson graphs and more so in the small illustration on page three of the paper—the peek into the future world of fractals, seeing a fractal before it was named so.

Richardson (a kind of Jevons, Marey-like American figure on the history of 20^th century science) did some very good thinking about the elusive nature of naming such a thing as the lengths of coastlines and other natural constructions: namely he determined that their measured length was dependent upon the scale of measurement. That is to say the measured length, L(G), is ruled the scale of G, approximating the function as L(G)=M(G)1-0 

Mandelbrot moves on to discuss the coastline in terms of it being a self-similar object, which is something that is nearly exactly similar to a part of itself, or the whole having the same shape of one or more than one of its parts, and regresses.

 

 

 

The Mandelbrot Fern (courtesy of http://en.wikipedia.org/wiki/Image:Fractal_fern_explained.png) displaying self-similar properties. 

An Outline Map of a Great Complexity: the ENIAC, 1948.

"The ENIAC exists independent of its history"--Brainerd and Sharpless

J.G. Brainerd & T.K. Sharpless. "The ENIAC."  Article occupying pp 163-172 of the February 1948 issue of Electrical Engineering.  Lovely copy, one old ownership stamp at top right.  $750

The ENIAC (Electronic Numerical Integrator And Computer, b. 1943-45) was, basically, the world’s first operational, high-speed digital computer, and the father of the computer industry.  What we see to the left  is the floor plan for the computer-with-no-monitor—I know to most people working today with a computer that the idea of a “floor plan” for anything that is not in a dark place at NSA is not easily conceivable.  The 30-ton, 18,000 tube, 125 KHz ENIAC’s space was about 1800 square feet, where it was able to add about 5000 numbers/second, which was vastly faster than anything else in existence. It operated with 70,000 resistors, 10,000 capacitors, 5 million hand-soldered joints and 6,000 manual switches. It was a magnificent achievement.

There are many sites dedicated to the history of the ENIAC—all I want to do here is surface this detailed floor plan of the computer, which seems not to be online.  The diagram comes from an article by J.K. Brainerd and T.K. Sharpless (Sharp brain!), “The ENIAC”, from the February 1948 issue of Electrical Engineering, which defines the machine and its components in full.  The abstract states: “The ENIAC is the only electronic large-scale general-purpose digital computing device now in operation.  Its speed of operation compares favorably with other electric and mechanical computers.  Developed under wartime pressures, it has been of value not only in producing results but in pointing the way toward improvements for future designs.”  Indeed.  The authors, both of whom worked on the construction of the machine from 1943 onwards, describe it in the following ten points:

“1. Large-scale 2.General purpose 3. Digital 4. Electronic  5.  Uses 10-digit numbers   6. Uses decimal system   7. High speed   8. Synchronous operation    9. Some parallel operations possible 10. Complete flexibility within limits of programming capacity” (page 170).

The authors make the following (qualified) understatement: “electrical engineers in the United States
have had a major interest in development of large-scale computing devices” mentioning the major players as MIT, Moore School (U PA), GE, Bell Telephone, Harvard and the spelled-out “International Business Machine corporation”.  Within ten years of the end of WWII the second generation of computer was nearly functional, and (as can be seen in the wonderful computer tree below) there were dozens of new machines in operation. By 1960, the explosion was getting ready  with the advent of microminiaturization. 

The computers in operation from 1938 to 1948 include the following alphabetical hodgepodge: Z1 Z2 Z3 Z4, ABC, Bell Model II, S1, Mark I, Mark II, Elliot 152, ARC, Mark II, SSEM, and the more poetic Colossus and Demon. (Zuse Z1 1938;  Atanasoff ABC 1939;  Zuse Z2 1939;  Bell Labs Complex Calculator aka Model I1940;  Zuse Z3 1941;  Zuse S1 1942; Bell Labs Model II; 1943 Bletchley Colossus Mark I 1943; Zuse S2 1943; Bletchley Colossus Mark II 1944; Harvard Mark I 1944; Zuse Z4 1944; Bell Labs Model IV  1945;  Bell Labs Computer Model V  1947;  Elliott 152 1947; Elliott 153 1947;  Moore School ENIAC 1947;  Birkbeck ARC 1948; Harvard Mark II 1948; Manchester University SSEM 1948;  NSA DEMON 1948.)

Quality Control of America's Anti-Bomber Defensive "Shield": the SAGE System, 1958

Luster, P(eter) K.  Quality Control of a Large-Scale, Real-Time Program System

Dated 7 August 1958. 11x8 inches. First page is an onionskin carbon of an abstract for the paper, followed by 11 sheets of offset, double-spaced text, with numerous author's corrections and additions.  About 3,500 words, total. RARE.  Unlocated.   $1500.

This is a proposal for selecting the proper way for a computer--in this case the SAGE computer and System--to monitor itself in real time and make assessments of its own reliability.

The SAGE  (The Semi-Automatic Ground Environment) was an air traffic control device that was used to detect enemy aircraft, relate their position and speed to interceptor planes who could then shoot down the threat.  It was basically the air defense shield against attack of North America by Soviet bombers (in the pre-ICBM days of MAD). The MIT/IBM SAGE system was enormously important in the history of computing because of its many real-time, interactive and online aspects--it was an enormously successful and futuristic system, one of the "best" computers ever built. 

From the paper:
"It has been necessary to device tests of the quality of the SAGE computer program in order to ensure its acceptability prior to installation and operation...also to run it "live" in order to give an immediate indication of any degradation of the overall system.  In effect, allow the SAGE computer itself to decide the quality of the SAGE computer program...The discussion here is limited to examining the nature of the problem to which such techniques are to be applied, and showing how the application may be made most effectively."

Luster (who at one time worked for a contractor in Arlington, Virginia, named the John D. Kettelle Corp) presents a "pre-statistical" account of the issue of the correctness of the real-time information that was being generated and produced by SAGE. 

A History of SAGE from WIKI:

(The problem of protection from what had previously been seen as "immune" strategic bombers was acute.) "It was this problem that particularly bothered Dr. George E. Valley, an MIT physics professor. In order to provide any sort of protection for the entire US, a series of radar stations would have to span both coasts and across Canada. In the event of a raid, there would simply be far too many reports to be able to successfully guide interception. His solution was automation, connecting all of the radar sites to a computer which would then control all of the incoming and outgoing flow of information. The interception operator's workload would be greatly reduced; they simply had to tell the computer which targets to attack, and perhaps choose what assets to use. All of the communications would be handled by the computer, and would be effectively instantaneous."

"This would require the system to update the operators in real time, and the only system in the world capable of doing this in 1948 when Valley studied the problem was the Project Whirlwind computer at MIT. The Whirlwind project, originally intended to control a US Navy flight simulator to train bomber crews, had run into problems and the Navy was losing interest. Valley talked to Jay Forrester, leader of the Whirlwind project, and together they wrote a study proposal to use Whirlwind for air defense."

"The United States Air Force was interested, and in 1949 they provided funding under the name Project Charles to develop a demonstration system. Information from several radars in the Cape Cod area was forwarded to the Whirlwind, which then developed tracks for the targets being reported. The "Cape Cod System" was a qualified success, and the Air Force took over the project under Project Claude, moving development to the new MIT Lincoln Laboratory in 1954. Making a military-grade version of the Whirlwind was a massive project that required close connections between Lincoln Labs, industrial partners who would build the machines and communications, and the military. In order to provide oversight and management during the deployment phase, MITRE was formed in 1958 to take over the project."

"Production of the resulting machines, known technically as the AN/FSQ-7, was initially awarded to RCA but later given to IBM, who started production in 1958. The buildings and internal power supply communications, and the 24 sectors' systems integration and test were provided by Western Electric, phone lines by the Bell System, and the software, 500,000 lines of assembly language, by a spin-off of RAND Corporation called System Development Corporation (SDC)."

On Reliability and Quality Control:

"The AN/FSQ-7 is the largest computer ever built,and will likely hold that record in the future. Each machine used 55,000 vacuum tubes, about ½ acre (2,000 m²) of floor space, weighed 275 tons[1] and used up to three megawatts of power. Although the machines used a large number of vacuum tubes, the failure rate of an individual tube was low due to efforts in quality control. Each SAGE site included two computers for redundancy, with one processor on "hot standby" at all times. In spite of the poor reliability of the tubes, this dual-processor design made for remarkably high overall system uptime. 99% availability was not unusual."