Skip to content

Gary Starkweather

Abstract

Gary Starkweather invented the laser printer at Xerox, and he had to fight his own company to do it. In 1969, working at Xerox’s research center in Webster, New York, he proposed pointing a laser at a copier drum to print images sent from a computer. His managers thought it was a waste, worried the laser would ruin the expensive drums, and threatened to break up his team. He escaped to the newly formed Xerox PARC in 1971, built a working machine there, and produced the technology behind the Xerox 9700, the one PARC invention Xerox actually turned into billions of dollars while it let the mouse and the graphical interface walk out the door. Starkweather later did color work at Apple and display research at Microsoft, and won a technical Academy Award for scanning film into computers.

The Optics Man at a Copier Company

Starkweather was born on 9 January 1938 in Lansing, Michigan. He took a physics degree from Michigan State in 1960, worked at the lens maker Bausch & Lomb, then joined Xerox in Rochester, the company built on xerography, Chester Carlson’s process for copying documents with static electricity and toner, and completed a master’s in optics at the University of Rochester in 1966 while working there. Xerox owned the physics of putting an image on a drum and fusing it to paper. What it did not yet have was a way to feed that drum an image from a computer instead of from a lens pointed at an original.

Starkweather’s idea was to supply the image with a laser. A laser beam, switched on and off and swept across the spinning drum, could write any pattern of dots the computer described, letters, diagrams, halftones, anything, at the full speed and quality of the xerographic engine. It would make a printer that was not tied to a fixed set of typebars or a photographic master. In 1969 that was a strange thing to propose at a company that sold copiers.

Fighting His Own Employer

The resistance came from inside. Starkweather’s division was in the business of copiers, and to the managers there a laser was an expensive solution looking for a problem, one that might burn out the drums they depended on. He was told the project was a dead end. His team was cut, and he was pressed to work on something conventional. By his own account he was close to being shut down entirely.

What saved the idea was timing. Xerox was setting up a new research center in California, the Palo Alto Research Center, to think about the “office of the future” without the copier division looking over its shoulder. Starkweather engineered a transfer, moving across the country in 1971 to a place that would let him build the thing his own colleagues had tried to kill. It is the same institution that produced the graphical user interface, the mouse, and Ethernet (see The Xerox PARC Revolution), and Starkweather’s printer was one of its first working results.

Building It at PARC

At PARC, with the freedom and the budget the Webster labs had denied him, Starkweather built a laser scanning head onto a Xerox 7000 copier and demonstrated laser printing in 1971, printing computer-generated images by 1972. He called the device SLOT, the Scanned Laser Output Terminal. Paired with a character generator, it became EARS (Ethernet, Alto, Research character generator, SLOT), and it hooked directly to PARC’s Alto personal computers over Ethernet. For the first time a researcher could compose a page on a screen, in multiple fonts and with graphics, and print it at high resolution on demand. That combination, a personal computer, a network, and a laser printer, was the office of the future PARC was chartered to imagine, working years ahead of the market.

The research prototype became the Dover printer used inside PARC, and the technology grew into the Xerox 9700, released in 1977 as a fast laser printer for corporate data centers. IBM had beaten Xerox to a commercial machine with the 3800 in 1976, but the 9700 line built on Starkweather’s design became a large business.

The Invention Xerox Did Not Fumble

Xerox PARC is famous as the lab that invented the future and let others sell it. Apple built the Macintosh from PARC’s graphical interface; 3Com sold PARC’s Ethernet; the mouse became everyone’s. The usual verdict is that Xerox profited from almost none of it.

The laser printer is the exception. High-speed laser printing became a business worth billions of dollars to Xerox itself, one of the few PARC ideas the company managed to commercialize on its own and the one that arguably paid for the whole experiment. The mass desktop market did go elsewhere: Canon built a cheap print engine that let Hewlett-Packard ship the LaserJet in 1984 and Apple the LaserWriter in 1985, and the LaserWriter’s PostScript (see John Warnock and PostScript) touched off desktop publishing (see The Desktop Publishing Revolution). Nor was laser the only branch: the machines most people actually owned hit the paper with pins or squirted ink at it (see The History of Printing). But at the high end the machine Starkweather had to smuggle to California earned back its cost many times over. The invention his managers tried to kill became the thing that justified the lab they never wanted to fund.

After Xerox

Starkweather left Xerox and joined Apple in 1988, where his optics background moved him into color. He led the development of ColorSync 1.0, Apple’s system for making a color look the same on a scanner, a screen, and a printer, and received the Optical Society’s David Richardson Medal in 1991 for contributions to optical engineering. In 1994 he shared a technical Academy Award, with Gary Demos, Dan Cameron, David DiFrancesco, and Scott Squires, for pioneering work in film input scanning, the step that gets motion-picture film into a computer and made digital effects and compositing practical. The Academy’s citation lists him “of Pixar”, which is a detail worth unpicking: he was at Apple by then, and the scanning work itself was done for Lucasfilm while he was still at Xerox, starting with consulting on the digital effects of the original Star Wars. Pixar is where that Lucasfilm graphics group ended up after 1986, and the award followed the work rather than the payroll. He joined Microsoft Research in 1997 to work on display technology, was elected to the National Academy of Engineering in 2004, and retired in 2005. He died of leukemia on 26 December 2019 in Orlando, Florida.

His career kept the same shape it started with: an optics specialist who kept finding the place where light, a computer, and an image meet, and who is remembered above all for the afternoon in 1969 when he decided a copier should take orders from a laser, and refused to be talked out of it.

📚 Sources