Who Invented the Computer? Gary Starkweather and the Laser Printer

Who Invented the Computer? This is the thirty-third installment in our ongoing series.

The laser printer may have been revolutionary, but Xerox couldn't be bothered to market it.

As soon as the first working computer was turned on, operators realized they would need an output device in order to read and save hard copies of their data. The immediate solution was to modify the teleprinter machines used by telegraph operators.

Oddly, in the rush to create ever more powerful computers, development in printing was seemingly an afterthought. For decades, the printer remained the slowest component of the computer system and, because it was entirely mechanical, often the most fragile and prone to breakdowns.

This industry-wide disregard for printer development wasn’t a problem, since early computers functioned primarily as powerful number-crunchers. Hard copies of the data produced didn’t require much in the way of printing power or finesse.

Although improvements were made in printers, they remained what they had always been: impact printers, so named because they operated much like a typewriter bringing an ink ribbon into direct contact with paper. Examples of these types of printers included dot matrix, daisywheel, and ball printers.

Impact printers were adequate for printing reports consisting of numbers, letters, and basic graphs. Users did, however, have to deal with a number of operational issues such as slow print speeds, ink ribbons slipping or wearing out, mechanical jams, and aggravatingly loud noises whenever these devices were in operation.

As computer use became more commonplace, both in the workplace and at home, powerful programs were created that included complex charts, diagrams, and even photographic images. It was time for a better printer.

Computer manufacturers were soon experimenting with ways to increase print speeds and improve image quality. Some of their initial ideas involved non-impact devices that utilized heat and electrical phototypesetting.


In 1960, the XEROX corporation released its 914 plain paper photocopier. It was an immense success, capable of copying originals that fit at or within the 9-inch by 14-inch window (hence the name) and printing out an impressive seven copies per minute.

The 914-copier operated in a surprisingly simple manner: a cathode ray tube (CRT) would shine a bright light onto a mirror that reflected the image of an original document onto a photosensitive rotating drum creating a negatively charged image on the drum. As the drum rotated through the toner tray it would attract positively charged toner particles as it passed.

A piece of paper, simultaneously fed under the drum, would receive the positively charged toner particles from the drum, creating a copy of the original document. The copy would then pass through a ceramic fusing element to melt the toner onto the paper and an exact replica of the original document would roll out of the machine.

914s were soon operating in offices everywhere. During its 17-year production, according to Fortune Magazine, the 914, measured by return on investment, is “the most successful product ever marketed in America.” A model of the 914, number 517, rests in the Smithsonian Institution in Washington, D.C.

Man with a Plan

The laser printer may have been revolutionary, but Xerox couldn't be bothered to market it.

In 1967, Gary Starkweather, a XEROX engineer was toiling away in the company’s research facility in upstate New York. While tinkering with the 914, he decided to remove the CRT and the original document and instead use a laser, modulated by a computer, to scan the image on a computer screen directly onto the rotating drum.

Starkweather's idea worked. He created a copy without using a physical original and, in the process, created a new output device for computers. Starkweather immediately approached his superiors about his breakthrough and requested additional resources to further develop it.

Somewhat surprisingly, the higher-ups didn’t view Starkweather’s project as a breakthrough. They reminded him that XEROX made copiers and ordered him to work on projects related to copiers.

For three months, Starkweather continued working in secret attempting to refine his prototype. When regional management became aware of his actions they threatened termination if he didn’t stop and get back to work on copiers. Unwilling to lose his job, Starkweather reluctantly set his project aside.

Shortly afterward, while reading a company newsletter, he learned that XEROX was establishing an additional research center in Palo Alto, Calif. The sole purpose of the new facility would be to develop devices that would operate in “The Office of the Future.”

Starkweather quickly wrangled a visit to the new Palo Alto Research Center (PARC) and convinced its managers to let him join the team. He arrived in Palo Alto in 1971 and, nine months later — to the great delight of PARC management — had a working laser printer.

Sell No, We Won't Go (to Market)

Once again, however, XEROX refused to market Starkweather’s laser printer. The company had been dabbling in electronic printing and sales and marketing teams were busy selling the impressive new 3600 Optical Character printer.

One year later, management did give PARC permission to ship out two Xerox Graphics Printers (GP) for trial use. One model was sent to the Stanford Research Center, while the other went to the Artificial Intelligence Lab at MIT.

The GP printed 180 dots per inch on a roll of 8.5-inch paper that the machine would cut into single pages during operation. While no GPs were ever sold, Stanford and MIT both reported being highly satisfied with the machines.

By 1976, Starkweather and his PARC colleagues created an improved laser printer, the XEROX Dover, able to print 2 pages per second at 300 dots per inch (dpi). PARC built 35 copies of the Dover, but XEROX again refused to market them, permitting them to only be given to universities.

Never Say Never

The laser printer may have been revolutionary, but Xerox couldn't be bothered to market it.

Finally, by the end of 1976, XEROX managment realized that the company needed to start selling laser printers when IBM began marketing the IBM 3800. XEROX would soon counter with the powerful XEROX 9700 laser printer that printed 300 dpi and up to two pages per second on cut-sheet paper.

As other companies like Hewlett Packard and Ricoh began manufacturing printers, new printer models improved dramatically, becoming much faster while simultaneously dropping in price and decreasing in size.

Laser printers have since become ubiquitous in most businesses and in many homes. These machines are now reasonably inexpensive and capable of printing up to 100 pages per minute at 600 – 1200 dpi. All because of Gary Starkweather (who died in 2019, at age 81) — a determined visionary who refused to give up on his idea.

Would you like more insight into the history of hacking? Check out Calvin's other articles about historical hackery:
About the Author
Calvin Harper

Calvin Harper is a writer, editor, and publisher who has covered a variety of topics across more than two decades in media. Calvin is a former GoCertify associate editor.