Выбрать главу

Drinkwater had almost become an IBM man but a couple of his old Navy buddies had been headhunted by Burroughs, so he had turned his back on the Big Blue, not knowing that in years to come he would be working hand in glove with the very corporation that the American public regarded as his company’s biggest, most ruthless competitor.

By the late 1950s there were nine great US computing powerhouses: IBM was the biggest by a distance but the other eight were all world players and market leaders; Burroughs, Honeywell, NCR (National Cash Register), General Electric, CDC (Control Data Corporation), RCA (Radio Corporation of America), Sperry, and DEC (Digital Equipment Corporation). By the dawn of the 1960s IBM’s market position had seemed so dominant that computer industry insiders — who knew well enough to leave Burroughs out of the equation — had begun to refer to ‘IBM and the seven dwarves’ to describe the unquestioned ascendancy of International Business Machines in global computing.

However, what the man in the street did not know, but what many in corporate America and elsewhere in the West suspected, and as time went by came to understand and to rue, was that IBM’s and the entire US computer industry’s research, development and core advanced technology production had been, ever since 1945 almost wholly underwritten by the US Department of Defence. The plain fact of the matter was that the mammoth scale of that support in the form of mind-bogglingly lucrative contacts — year after year — coming out of the Pentagon had been so vast, and the political gerrymandering behind the open-ended subsidies priced into those contracts so complex and so gross, that not even IBM could think of ways to spend all the monies that had poured like great rivers in flood into its coffers; hence the Burroughs Corporation, and every one of the other ‘seven dwarves’ had also grown fat on the Government’s largesse via huge, often long-term IBM-managed and or, contracted out projects.

At the heart of the unstoppable, meteoric rise and rise of the American computing industry, remorselessly fuelled with ever more urgency and well, paranoia, during the 1950s as successive Administrations battled to close first the non-existent ‘bomber gap’, the equally imaginary ‘missile gap’ and then to combat the highly embarrassing apparent — and very public — propaganda nightmare of the Soviets gaining a march in the ‘space race’, was SAGE. The acronym SAGE — the letters standing for Semi-Automatic Ground Environment — described a system comprising tens of large, so-called mainframe computers, and all the hard-wired networking equipment and communications infrastructure required to co-ordinate data inputs from all connected radar and related intelligence resources; thereby to enable NORAD to detect, analyse and control its response to a Soviet air attack on the North American continent. SAGE was a rapacious monster which had by the early 1960s sucked up and greedily consumed a budget that made the cost of the Manhattan Project look like loose change, so much petty cash.

It had only cost the American tax payer a piffling $2 billion to build the atomic bomb; IBM had been handed $10 billion — and change — to create and to implement SAGE. Leastways, that was what Carl Drinkwater had been told by his boss; and why would the man lie to him about a thing like that?

Back in 1949 Carl Drinkwater had had no idea what he was actually working on, and nobody at Burroughs with the necessary security clearance had gone out of his way to explain. Burroughs had still been called the Burroughs Adding Machine Company in those days, not becoming a ‘Corporation’ until 1953. However, Carl had known the company was working on ‘something big’ and on a number of ‘top secret military projects’; he had not gone to Caltech and discovered the ‘God’ of the natural universe just to spend the rest of his working life designing and building better and bigger ‘adding machines’. What he had not known, and what he would not have believed had he been told back then, was the incredible, breathtaking scope and ambition of SAGE.

When the Headquarters of NORAD at Ent Air Force Base located at Colorado Springs became operational in 1957 it was at the hub of a system of nearly two hundred radar stations and regional ADCs — Air Defence Centres — covering the entire North American continent. Each ADC was a giant four-storey concrete blockhouse with a ground footprint large enough to accommodate a football pitch, hardened against over-blast pressures of up to five pounds per square inch, which accommodated not one but two one hundred and thirty-five ton IBM-Burroughs mainframe computers, and was responsible for its own designated airspace defence sector.

Notwithstanding that SAGE was still a work in progress, since it had become operation in 1957 most of the envisaged two dozen ADCs — great reinforced concrete blockhouses like the one at Ent Air Force Base with identical equipment and communications inputs and outputs — had been completed and the newest additions to the network were in the process of final commissioning. Each individual ADC was a monolithic marvel of American applied science and unambiguous symbols of unrivalled technological might. The two great computers within every one acre-sized four storey concrete ADC each took up 7,500 square feet of floor space, mounted 60,000 vacuum tubes, 175,000 diodes and 13,000 transistors and incorporated a seemingly astonishing 256 kilobytes of magnetic core or, as it was increasingly referred to, Random-Access Memory. Carl Drinkwater found himself staring dreamily into thin air every time he thought about that amount of magnetic core RAM. The British, who had invented the first true electronic programmable computers to break the German Enigma code in the Second World War were still tinkering with improved versions of Colossus, the first 1940s Bletchley Park code-breaking machine; nobody else in the World had anything remotely like SAGE. Each SAGE mainframe computer consumed three megawatts of electricity and ran so fast and got so hot that each ADC kept the second mainframe at immediate readiness to take the load if the first one crashed. Connected to the other ADCs with unbelievably fast state of the art top secret modems — capable of a lightning 1,300 baud data transfer rates — and able to consistently execute up to 75,000 instructions per second, SAGE could literally churn out more data than the ADC had cathode ray tubes on which to display it!

Every time Carl Drinkwater walked into the control room he felt like a character out of a science fiction novel transported in the blink of an eye by some magical time machine into the far distant future. Now and then he allowed himself to speculate — whimsically — if, assuming the current trend towards miniaturisation continued whether in ten or twenty years time a machine with super-advanced transistors and circuit boards might one day be so small, that computers as stunningly capable as the SAGE mainframes would fit into a small room, or even a box on a desk in his office?

God in heaven, science was great!

Carl tried not to daydream when he was at work.

Every output from the other ADCs fed back into the control room at Ent Air Force Base via a hardened network of AT&T — American Telephone & Telegraph — dedicated lines and modems in real time. The air defence controllers manning the serried ranks of gun metal consoles stared constantly at their flickering big round cathode ray tubes. At any time individual displays could be projected singly, or in combination onto the big, backlit wall projections of the North American continent. SAGE had never been just a computer project. To make it work, to actually make it useable research in countless related fields had been lavishly funded, university research departments and American business had been showered with money to investigate obscure areas of apparently ‘pure science’, often with no idea what they were involved in, nor the least inkling what possible purpose their work might serve. Out of the mass of scientific and technological discovery, and the frenzy of development and innovation that SAGE had spurred had emerged vastly improved transistor technologies, new microwave radio applications, countless quantum leaps and breakthroughs in data handling, storage and analysis, and ever better and more efficient means of displaying images, data, and giant strides forward had been achieved in the transmission and reception of television pictures. Some of the fruits from the ongoing development of the core SAGE technologies and from the myriad of related projects, were already fuelling a new American high-tech manufacturing boom. But all that was ephemeral to the mission of NORAD.