Welcome back to Input Origins, our monthly time machine through the evolution of how users control machines.
This is our 20th edition, and to mark the milestone, we're telling the story of the input device that birthed the graphical user interface. The one that let us point at a screen, for the first time, and get an instant response.
The Light Pen.
And although the light pen was the firing shot for the age of graphical interfaces, that same interface is what had rendered it obsolete. A testament to how input methods come and go, yet the paradigm underneath prevails.
Series Editor: Ariel Amar
The first light pen, named "Light Gun". Used for the the SAGE System (1950s).
Born in the Cold War
The state of computer input in the mid-1950s meant that interacting with a computer entailed feeding it a stack of punch cards and waiting hours for the output slip (waiting too long for that prompt to come through, anyone?).
This was also the Cold War era, the US Air Force was watching the skies for Soviet bombers. Operators needed a way to point at a hostile blip and have the computer respond IMMEDIATELY, yet the only input the computer understood was a punch card.
Come Robert Everett, of the MIT Lincoln Lab. He had the ingenious idea of putting a photodiode in a tube wired to the computer… in the shape of a gun. Once the operator pointed it at the CRT radar-monitor, Everett's gun would detect the exact XY coordinates the CRT monitor's beam was pointing to, instantly transmitting these onwards.
Everett's invention catalyzed one of the most profound paradigm shifts in computing history: transforming the computer from a text-processing math engine into a responsive, visual extension of human cognition.
Light-gun–operated radar screens used by NORAD
It Almost Escaped the Lab
So did it end up in home use? Eventually, yes! But first it spent twenty years inside the most expensive enterprise computers on Earth. The IBM 2250 Graphics Display Unit shipped with a light pen and a price tag of $80,000 per terminal. For two decades, interactive computer graphics was synonymous with the light pen, across aerospace, automotive, medical imaging, oil & gas seismic mapping, and military command consoles.
Then, briefly, the light pen came to personal homes. Between 1979 and 1984, almost every home-computer maker shipped or licensed a pen, Fairlight, Atari, the Commodore 64. But it was alas brief, Atari pulled its pen from production within months while the others quietly faded. The consumer light pen was a brief, optimistic, slightly clumsy moment. So what killed it?
Steve Gibson's Light Pen System II for the Apple II (1982), later distributed by Koala Technologies.
The "Gorilla Arm" Syndrome
Try holding your arm extended horizontally, holding a pen to point at a vertical monitor for an eight-hour shift. Within minutes, your shoulder burns, your bicep fatigues, and your hand begins to go numb.
The HCI community calls it "Gorilla Arm". While the pen answered our innate need to interact with the world physically (i.e., the point-and-click paradigm), the computer mouse delivered this strain-free, by decoupling the physical position of the hand from the visual cursor.
But while the mouse pushed the light pen out of the office, the transition to LCD monitors delivered the final, fatal blow. As we mentioned prior, the light pen relies on the flash of the CRT monitor's beam. LCDs operate in an entirely different manner, no scanning beam meant no pulse for the pen's photodiode to detect, which made this technology physically impossible to use.
Extended use of mid-air interfaces can lead to arm fatigue. (Purdue University Photo/Brian Huchel)
Gorilla Arm 2.0 for AR?
The physical pen died, but the paradigm it pioneered keeps reincarnating to this day. It paved the way for the touchscreen, proving that our primal urge to reach out and physically touch the digital world is ever-present.
As we enter the era of AR and spatial computing, are we blindly repeating history? Computer vision tracks our hands in mid-air, forcing us to swipe and pinch at invisible floating menus. Welcome back, gorilla arm. (History, it seems, loves a punishing shoulder workout).
We still want to point and pinch. We just want to do it from comfortable body postures. EMG neural input supports the same familiar hand gestures we already use with camera-based systems. And since the hands can rest naturally along the waist, it avoids the gorilla arm effect caused by camera line-of-sight and field-of-view constraints. The result is intuitive spatial control without gorilla arm 2.0.
Even Tom Cruise needed breaks filming Minority Report’s mid-air gestures
