In 1956, an anonymous IBM employee made a lady from the pages of Esquire come to life on the screen of a $238 million military computer.
Pin-up program running on an SD Console (Lawrence A. Tipton)
During a time when computing power was so scarce that it required a government-defense budget to finance it, a young man used a $238 million military computer, the largest such machine ever built, to render an image of a curvy woman on a glowing cathode ray tube screen. The year was 1956, and the creation was a landmark moment in computer graphics and cultural history that has gone unnoticed until now.
Using equipment designed to guard against the apocalypse, a pin-up girl had been drawn.
She was quite probably the first human likeness to ever appear on a computer screen.
* * *
In the early years of the Cold War, the United States invested billions of dollars into a state-of-the-art computerized air defense system unlike any the world had ever seen. At the time, the threat of nuclear strike came primarily from Soviet bombers penetrating American airspace, so the Air Force, in conjunction with MIT and IBM, created SAGE: Semi-Automatic Ground Environment. It was a project so ambitious that it ranks among the massive scientific programs like the Manhattan Project and Apollo in scope and budget, yet surprisingly few people know about it today.
The SAGE system defended against Soviet attack by combining live radar input with pre-programmed commercial airline flight information to paint a live picture of what should or shouldn't be flying in American airspace. If something looked out of place, SAGE commanders could scramble jets to intercept the unknown object.
Starting in 1958, the Air Force built 21 SAGE centers around the continental U.S. At the heart of each center lay an enormous, 4-story windowless blockhouse that contained two massive AN/FSQ-7 computers. Each computer cost the equivalent of $1.89 billion in today's dollars and occupied half an acre of floor space.
A typical SAGE blockhouse schematic
The AN/FSQ-7 was an achievement unto itself, for it was the second real-time computer with an electronic graphical display in history. The first was the Whirlwind I, the 1950 precursor to the AN/FSQ-7 that began as an MIT project commissioned by the Navy for flight-simulation purposes.
Computer frames of the AN/FSQ-7 Computer System, 1958 (MITRE)
At any given time of the day, technicians at each SAGE center kept one of the dual AN/FSQ-7 computers live while the other underwent maintenance. That way, even in an era of unreliable vacuum tubes (of which each computer contained 50,000), SAGE could maintain near 100 percent uptime.
At the top story of every blockhouse lay dozens of OA-1008/FSQ situation display consoles (91 in a typical installation), each hooked to the active AN/FSQ-7 computer, upon which personnel would monitor different segments of airspace in a given defense sector. For example, the SAGE center at Ft. Lee Air Force Station in Virginia was responsible for monitoring a sector which included the Washington, D.C. area.
Each situation display console in the blockhouse contained a specially crafted 19-inch cathode ray tube (CRT) display that could draw vector-based lines or alphanumeric characters on any portion of the screen. One may be familiar with CRTs from their use in television sets, but unlike a television set, the CRT in a situation display console did not paint a top-to-bottom raster-based image on the screen. Instead, it drew lines from any point to any other point arbitrarily, sort of like an electronic Etch-a-Sketch.
A typical SAGE situation display
Typically, the SAGE computer drew a line-based image of coastline or map on each SD console then overlaid flight vector information and live radar blips. When the operators wanted to identify a certain flight on the screen, console operators had a unique input device at their disposal: the light gun. They simply pointed the light gun at the spot and pulled the trigger. Alphanumeric information that identified the flight would then appear next to it on the screen.
A SAGE operator at a Situation Display Console, holding a light gun (IBM)
Among this very serious business of detecting incoming nuclear threats, someone decided to commandeer the AN/FSQ-7's computing power and graphical display for something a little more fun. Using a console and one of the two SAGE computers, an enterprising programmer created a graphical interpretation of a pin-up style woman that rendered as line segments on the SD's 19-inch screen. In doing so, the pin-up's programmer created the world's earliest known figurative computer art, and quite possibly the first image of a human being on a computer screen.
In early 1959, 21-year-old Airman First Class Lawrence A. Tipton snapped the only known photo of this pin-up program in action at Ft. Lee. The photo shows the tube of an SD console displaying the outline of woman with her arms held high, cradling her head while emphasizing her bosom. She reclines awkwardly, her legs splayed apart in an uncomfortable but provocative pose that smacks of mid-century pin-up art.
"One day I decided to take pictures for posterity's sake," recalls Tipton, "And those two Polaroids are the only ones that made it out of the building." The other Polaroid is a self-portrait of Tipton himself sitting in front of the AN/FSQ-7's Duplex Maintenance Console. "We used the Polaroid cameras to take pictures of anomaly conditions. When the computers would malfunction, you'd take pictures of those main consoles to diagnose the conditions."
Lawrence Tipton in front of the Duplex Maintenance Console (Lawrence Tipton)
The pin-up image itself was programmed as a series of short lines, or vectors, encoded on a stack of about 97 Hollerith type punched cards, Tipton recalls. Hollerith punched cards were 7.375 x 3.25 inch paper cards that stored binary data via holes cut through a matrix printed on its surface. Like other 1950s computers, the AN/FSQ-7 used the cards extensively for program input.
A Hollerith punched card (IBM)
According to Tipton, the program that displayed the pin-up image was a diagnostic that tested data flow between the two SAGE computers on site (referred to as the A and B computers). At the end of every shift, as one computer was about to go offline and switch over to the other, the active machine would begin transferring flight and intercept data to the standby machine so there could be a seamless switch over.
Two switching consoles on site were used to handle this process. After running the diagnostic, Tipton describes, if the pin-up displayed correctly on the screen, then data was being transferred between the A and B computers correctly. If the image displayed improperly, then the technicians immediately knew there was a problem.
When loaded, the pin-up image would be visible in flashing pulses that synchronized system-wide with the incoming flow of real-time radar data. A long exposure on Tipton's Polaroid camera would have assured the steady image of the pin-up you see here. (The pin-up lady has a spot on her thigh because that is the center of the circular display, which is where the electron gun in the CRT naturally aims when it is idle.)
When contacted, dozens of SAGE veterans who worked in the program between 1958 and 1983 (the approximate lifetime of SAGE) recalled witnessing this pin-up program firsthand. Not surprisingly, the pin-up's role changed over the decades as the technological culture shifted around it. While Tipton insists the pin-up had a diagnostic purpose, SAGE operators in later decades remember it as a lighthearted way to pass the dull hours of the late shift when traffic was slow or the standby machine was not in service ("At no time was the primary mission of air defense compromised to my knowledge," wrote one veteran in an email to the author.)
* * *
Viewing this ancient digital artwork today, one naturally wonders who created it. "I remember at the time that everybody knew it was done by an IBM programmer," recalls Tipton. Robert Martina, a veteran of early SAGE installations in the 1950s, agrees with Tipton. "IBM guys were supposed to be so straight. They weren't," he adds with a laugh. But no one seems to recall who exactly at IBM created it.
Going by accounts from Tipton and others, the pin-up program likely dates from 1956 to 1958. The upper end of the year range, 1958, can easily be established because multiple eyewitnesses claim that the diagnostic was present when the first non-test SAGE site went live in New Jersey in early 1958. The lower end of the range, 1956, comes from a compelling piece of cultural evidence.
In 1955, famed pin-up artist George Petty resumed a relationship with Esquire magazine just before his retirement. He illustrated two calendars for the publication, one for 1955 and one for 1956. Each month's page came accompanied by a lushly illustrated and extremely scantily clad Petty pin-up.
Petty had a way of painting a woman by which she almost appeared nude if not for a sheer, skin-hugging fabric that obscured almost nothing. Such is the case in the December 1956 calendar pin-up, which leaves little besides the woman's mysteriously absent nipples to the imagination.
As it turns out, this illustration matches the SAGE pin-up almost exactly (as you can see in the comparison below), as if someone directly traced her outline and translated it into vector coordinates. It's likely that whoever created the pin-up image used a technique similar to those used to encode maps and coastlines into vector segments for display on the system.
SAGE pin-up compared with George Petty pin-up (Benj Edwards)
Since the Esquire 1956 Girl Calendar was likely printed in mid-late 1955 to prepare for the coming year, and that painting was created specifically for the calendar, the pin-up could not have originated from before that time. It may have even been created when the December 1956 image was hanging on the wall nearby.
As it happens, in 1955, IBM began educating specially selected Air Force personnel as part of the first SAGE training program at its facilities in Kingston, New York. It's possible that IBM instructors at Kingston devised the pin-up diagnostic an amusing way to immediately engage a group of all-male, usually 18-24 year old computer novices. But until more information surfaces -- or someone comes forward to claim credit -- the author shall remain unknown.
While no SAGE veterans contacted by the author remember seeing the pin-up girl at Kingston, they do remember seeing another risqué diagnostic program in use by IBM instructors around 1960. It featured the rough outline of topless hula dancer with a grass skirt. And it was interactive. (No photos of the hula girl are known to exist.)
According to one account, upon proper manipulation of switches on the console, the hula girl's hips would sway back and forth in an arc, synchronized to music playing on the console's buzzer. If the operator pointed his light gun at her navel and pulled the trigger, her skirt would fall off and the screen would go blank.
"That probably was the first pornographic show," jokes Robert Martina, a SAGE veteran who witnessed the hula girl in action. "As close as it gets. It was so innocent."
As with the pin-up program, early SAGE vets recalled the hula girl as a diagnostic, and later ones remembered it as a form of lighthearted amusement. Its program seemed to vary from site to site: some vets say the girl danced on her own without the use of switches, and others say she wasn't topless. It's likely that the program was modified many times over the years.
How the diagnostics were viewed also changed from site to site, era to era.
While Tipton and Martina recall viewing the diagnostics during the early setup of various SAGE centers, John R. Stephens, Jr., who worked in the SAGE program from 1962 to '64, recalls times when unofficial word would circulate among the operators that one of the two girl-based programs would be displayed on one of the consoles. As in: "They're going to run girley2 on console 4-1 around 0300 local."
(According to Stephens, SAGE operators referred to the pin-up program as "girley1" and the hula girl program as "girley2.")
While formal tests and drills were regularly announced over the PA system, news of the risqué diagnostics only spread by word-of-mouth. "The girley programs were never announced over the PA, which led me to the conclusion they were beyond normal procedures," says Stephens.
But even if the images weren't totally above board, they encountered little to no resistance among the ranks. "It was an all male enterprise at that time," recalls Tipton, speaking of the late 1950s. "There were a few women in the Air Force upstairs in the control center, but at that time of life, you know, there wasn't as much controversy."
More From The Atlantic