14 Failure or Success?
As impressive as SAGE with its marvelous duplex AN/FSQ-7 and AN/FSQ-8 computers was, there was always harsh criticism from various sides: Some claimed that SAGE was a waste of taxpayer’s money with all of its “gold plated” equipment, while others called it a complete failure, incapable of performing the air defense task. Still others marvel at the technologies developed for SAGE and think of it as a good investment in the future of a whole industry. The following sections will shed some light on both sides of the coin.
14.1 A failure?
In some respect, SAGE was outdated even after its commissioning in 1958 as the main threat of air defense had already shifted from long range bombers equipped with thermonuclear weapons toward Intercontinental Ballistic Missile (ICBM) installations. The first of these new weapons, the A9/A10 rocket, was designed in Germany beginning in 1940, although no actual working system was built and tested. Nevertheless, the idea of a long range rocket based weapon system was prevalent since the 1940s and in 1953 development of an ICBM capable of delivering a thermonuclear warhead to a long distance target began under SERGEI PAVLOVICH KOROLEV.519 This led to the R-7 rocket which performed its first successful test on August 21st 1957 and reached a distance of about about 8,000 km, way enough to reach targets outside the USSR. Interestingly, General HENRY HARLEY ARNOLD520 somewhat prophetically wrote in 1943:
“Someday, not too far distant, there can come streaking out of somewhere – we won’t be able to hear it, it will come so fast – some kind of gadget with an explosive so powerful that one projectile will be able to wipe out completely this city of Washington.”521
Even the following 1947 statement of Brig. General THOMAS SARSFIELD POWER,522 which turned out to be correct, did not neglect the looming threat of ICBMs:
“For the next 10 years, long-range air bombardment will be effected by means of subsonic bombers only.”523
So it was foreseeable even in the late 1940s that the long range bombers, which were the basis of definition of the air defense problem which in turn led to the development of SAGE, were soon to be history, replaced by ICBMs against which next to no countermeasure was possible. Accordingly, the often heard accusation that SAGE was outdated from its very beginning is not without a grain of truth.524
Even a conventional strike of the USSR, throwing a plethora of long-range bombers against targets in the United States, would have been outside the detection and processing capabilities of SAGE. Although there were numerous test raids flown during operational weapons tests and evaluations, the realism of these scenarios may be questionable. The reported kill percentages, always near 100%, were obtained under ideal conditions with no electronic counter measures by an enemy such as radar jamming and the like.525
Another weak point of SAGE was the decision to build the DCs and CCs above ground.526 This was due to a half-hearted attempt at cutting costs. Of course, structures like the SAGE blockhouses were easy targets and might also have attracted an explicit attack:527
“[. . . ] SAGE blockhouses were conspicuous structures that dominated the landscape wherever they were located. A mere handful of enemy missiles, therefore, could severely cripple US defenses against the manned bomber.”528
PAUL N. EDWARDS remarks with a rather grim sense of humor that
“[the] decision [of building the SAGE installations above ground] had only one possible strategic rationale: [Strategic Air Command (SAC)] intended never to need SAGE warning and interception; it would strike the Russians first. After SAC’s hammer blow, continental air defenses would be faced only with cleaning up a weak and probably disorganized counterstrike.”529
On the other hand, SAGE might not have been useless at all from a military perspective – there never was a preemptive strike of the USSR against the United States, the Cold War never became hot. So SAGE might have at least been a successful deterrence.
Nevertheless, a 1969 incident clearly demonstrated the limitations of SAGE as a whole and should become the beginning of the end: On October 5th, 1969, a defecting lieutenant from Cuba landed with his MIG fighter on Homestead AFB. As embarrassing as this alone would have been, it was even worse since only a short time ago the Air Force One had landed on Homestead AFB with the president of the United States aboard.530 This debacle turned into a severe blow against SAGE as congress decided to cut back financial funding substantially in the aftermath.
14.2 Success!
Despite the undisputable immense costs of SAGE, common estimates range from $ 4–$ 12 billion, with $ 8 billion being a probable value,531 SAGE turned out to be a fundamental part of our technological heritage. Nearly all following developments in computer technology were pioneered in AN/FSQ-7. SAGE, in fact, created what became a computer industry. The countless spin-offs from this giant undertaking are so invaluable that SAGE can hardly be seen as a waste of time and money from today’s perspective.
14.2.1 Hardware
From a computer architectural point of view, AN/FSQ-7 was the first large scale production machine clearly demonstrating the power of indexed addressing with its BPX instruction.532 In addition to this, AN/FSQ-7 pioneered the idea of a dedicated input /output control unit capable of transferring blocks of data without intervention of the central processor, predating today’s ubiquitous direct memory access533 controllers.
By far the most important asset from Whirlwind and AN/FSQ-7 was the development of reliable large scale core memory systems. For several decades to follow this solved one of the most important problems of computer technology. Prior to these developments, memory systems were large, clumsy, and error-prone, often determining the ridiculously short up-times of early stored program digital computers. The very idea of three-dimensional stacks of magnetic cores to store binary data boosted a whole industry and also paved the way for complex and memory hungry software systems, including compilers and interpreters.
The reliable transmission of digital data over cheap voice-grade telephone lines was also pioneered by SAGE. The transmission devices, forerunners of today’s modems,534 became a central contribution to the Bell System A-1 data system535 and clearly demonstrated the practicability of nation- and even worldwide computer networks:
“The government also bootstrapped computer communication with Whirlwind, an Office of Naval Research computer that processed telemetry data in real time at MIT. Whirlwind demonstrated the feasibility of real-time data communication over analog telephone lines [. . . ] After the Whirlwind R&D, the government turned to procurement with the Semi-Automatic Ground Environment (SAGE) system [. . . ] SAGE was the first computer network, growing finally to link Q-7s and Q-8s in 26 centers.536
The AN/FSQ-7/8 systems also were the first large scale systems relying on circuit standards being employed throughout the system, ranging from basic circuits and their packaging in easily exchangeable modules, to standard mounting frames, intercab-inet cabling etc. Another feature pioneered by these machines was parity checking throughout the machine, involving every internal data transfer. It should take years before commercial machines featured so-called fully checked designs.537
AN/FSQ-7 also pioneered the idea of marginal checking and, much more influential, that of duplexing computers to increase availability, becoming the forerunner of today’s cluster solutions. MORTON MICHAEL ASTRAHAN538 and JOHN FRANCIS JACOBS539 aptly note that
“[t]he SAGE system provides a demonstration of the kind of innovation that can be achieved when cost is secondary to performance. This kind of environment is difficult to create in a commercially oriented company, but SAGE provided the environment. Ambitious performance goals were met by the operational systems. Furthermore, as hardware costs dropped, most of the SAGE innovations became cost effective for the commercial market.”540
14.2.2 Graphics
Figure 14.1: Pin-up displayed on a SD tube at the Fort Lee DC (©Airman First Class LAWRENCE A. TIPTON, 1959)
It should take years for other machines to reach the graphical capabilities of AN/FSQ-7 and its associated situation displays. These marvels of technology predated all of the later vector graphics displays and pioneered graphical user interfaces, GUIs for short, with the development of the light gun.
Of course, the availability of such a computer system resulted in not too serious programs, too. The now famous pin-up shown in figure 14.1 is “[the] world’s earliest known figurative computer art, and quite possibly the first image of a human being on a computer screen.” According to
LAWRENCE TIPTON, the program generating the pin-up image consisted of about 100 punch cards.541 Research done by BENJ EDWARDS shows that the picture displayed closely resembles a pin-up drawn by artist GEORGE BROWN PETTY IV542 for a December 1956 calender page.
14.2.3 Software
Being the first large scale software development project, SAGE coined and formed a complete sector of technology. It not only demonstrated the overwhelming and always underestimated complexity of software development as such but also offered the first formalized techniques of dealing with this new degree of complexity. It gave a first foretaste of the so-called software crisis and would become catalyst of a new subfield in computer science eventually called software engineering.543
Apart from these indirect contributions to software technology, SAGE pioneered many ideas and techniques which we today take as granted, such as time-sharing, real-time processing, modular (top-down) system organization, table-driven software simplifying later modification and extension, centralized data structures,544 etc. Especially, these central data structures turned out to be quite inspiring:
“By the late 1960s, however, ‘data base’ was a common expression used in corporate computing circles, largely replacing the hubs, buckets, and pools in which data had previously been rhetorically housed. This term was imported from the world of military command and control systems. It originated in or before 1960, probably as part of the famous SAGE anti-aircraft command and control network. [. . . ] SAGE had to present an up-to-date and consistent representation of the various bombers, fighters and bases to all its users. The System Development Corporation, a RAND Corporation group spin-off to develop the software for SAGE, had adopted the term ‘data base’ to describe the shared collection of data on which all these views were based. SDC actively promoted the data base concept for military and business use. Its interest in general purpose data base systems was part of its attempt to find new markets for its unique expertise in the creation of large, interactive systems. [. . . SDC] identified ‘computer-centered data base systems’ as a key application of time-shared systems – hosting (in collaboration with military agencies) two symposia on the topic in 1964 and 1965. The SDC Data Base Symposia were crucial in spreading the data base concept beyond the world of real-time military contractors.”545
14.2.4 ATC and SABRE
Another direct spin-off of SAGE is the application of networked computer systems to air traffic control.546 The MITRE Corporation contracted with the Federal Aviation Agency in 1959 in a program called SAGE Air Traffic Integration (SATIN). The aim of this project was the development of a single, nationwide system for tracking and managing all aircraft in that nation’s airspace.547 This work was, of course, based on the early ATC studies performed by DAVID R. ISRAEL548 in 1950 and 1951 using Whirlwind. 549 Eventually, but only for a limited time, even one SAGE site550 became part of the Federal Aviation Administration (FAA) air route traffic control center.
Another direct spin-off of SAGE is the SABRE551 system – the first computerized reservation system developed by IBM. Due to an incredible coincidence, BLAIR SMITH, an IBM salesman, was sitting next to CYRUS ROWLETT SMITH, the CEO of American Airlines during a business flight in 1953. As they talked, the idea of using computers – back then still electronic brains for most people – for flight reservations etc. came up and IBM jumped at this opportunity to apply the know-how gained during the (still ongoing) development of the AN/FSQ-7 prototypes to a large scale business application. The first incarnation of this revolutionary system became operational in 1964 after some $ 40,000,000 had been spent on research and development.
14.2.5 SAGE in popular culture
Finally, the AN/FSQ-7 had and still has a remarkable impact on popular culture –its maintenance consoles, featured in the settings of more than 60 movies,552 are still epitomes of giant computers. It is quite remarkable that the first appearance of an AN/FSQ-7 console in a cinema movie was in 1964 in “Santa Claus Conquers the Martians”, at a time when all of the documentation regarding AN/FSQ-7 was still classified. At the time of writing this book, the last movie making use of these marvelous consoles was “Future Shock”, episode 22, season 1 of “Flash Forward”.
It may be safely assumed that SAGE, and especially AN/FSQ-7, inspired the creation of “WOPR”, the fantasy “War Operation Plan Response” computer featured in the 1983 movie “WarGames”. The final sequences of this movie show a game of Tic-tac-toe being played, which causes WOPR to realize that there would be no winner in a global thermonuclear war, thus saving the world. It may just be a coincidence, but A. ZABLUDOWSKY demonstrated a Tic-tac-toe playing program on Whirlwind in 1955. . . 553