Why Computers Aren’t Patented

David Kalat

Patent sharing, royalty revenue and the invention of the computer

The ENIAC was a big deal in its day, both literally and figuratively. The Electronic Numerical Integrator and Computer was the first electronic, programmable, general-purpose computer. The ENIAC represented the manifestation of ideas that computer science theorists from Charles Babbage to Alan Turing had dreamt of, but which now had come into the physical world of vacuum tubes and wires. Sure, Star Trek could claim to have “invented” the matter transporter, but those boasts ring hollow until someone actually makes it possible for Scotty to beam up. The ENIAC was a marvel of engineering, and the engineers who made it rightfully went looking for their reward for doing so.

Staking a Claim

By the fall of 1944, J. Presper Eckert and his student John W. Mauchly at the Moore School of Engineering at the University of Pennsylvania decided that their design work on the ENIAC and its successor the EDVAC (Electronic Discrete Variable Automatic Computer) was complete, even though construction on those machines continued. Before filing any claims, Eckert and Mauchly circulated a letter asking their fellow engineers if anyone had a claim to the invention. Receiving no answers, they went forward.

Eckert and Mauchly filed their patent application in 1947; they eventually were awarded US Patent Number 3,120,606in 1964. In the years between those two events, Eckert and Mauchly started their own computer company. The Eckert–Mauchly Computer Corporation quickly floundered, and they sold it to Remington Rand, which became Sperry Rand. 

Sperry Rand was delighted to obtain the pending patent for modern computer technology. Their attempts to corner the market on computers were not, however, the source of significant royalty revenue. Sperry Rand had entered into a patent-sharing agreement with IBM, which gave IBM rights to make digital computers. Because IBM manufactured the vast majority of computers in that era, few other companies were left to pay the 1.5 percent royalty Sperry Rand charged for licensing the Eckert-Mauchly patent. However, one company opted not to pay the royalty at all.

Get to the Courthouse in Time

When Sperry Rand sued Honeywell in 1967 for infringing the ‘606 patent, it discovered that Honeywell had reached the courthouse first (by minutes) to file its own claim against Sperry Rand for antitrust violations.

At first glance, Honeywell’s strategy seemed quaintly quixotic. Sperry Rand had already defended Eckert and Mauchly’s claims successfully in court against a challenge from Bell Telephone Laboratories. That win came in 1962, two years before the patent was formally awarded. 

The Honeywell lawyers had a steeper hill to climb. But they were determined to try. The judge in the 1962 Bell Labs case found that there was insufficient evidence of any “prior public use” of the ENIAC’s design. This was what Honeywell’s attorneys set out to address.

In the end, Honeywell’s arguments came down to two critical events. 

In 1941, Mauchly spent several days visiting computer scientist John V. Atanasoff and his associate Clifford Berry. During that time, Mauchly saw their prototype digital computer and reviewed some its technical documentation. Atanasoff and Berry intended to file their own patent, but the attack on Pearl Harbor in December of that year reoriented their attention, and they did not submit their paperwork in a timely fashion.

Then, in early 1945, mathematician John von Neumann visited Eckert and Mauchly’s lab to see early work on the EDVAC. Taken by what he saw, von Neumann wrote an article entitled “First Draft of a Report on the EDVAC,” which he did not publish but distributed to some associates. 

 I’ll Be the Judge of That

The litigation between Sperry Rand and Honeywell was heard in the 4th Division of the Minnesota District Court (No. 4-67-Civ. 138). The case spanned five years; the trial itself lasted over a year, from 1971 to 1972. When all was said and done, 77 witnesses had testified orally, and another 80 were represented through depositions. The case generated 50,000 pages of trial transcripts, with 36,000 documents entered into evidence. 

The court found von Neumann’s draft article to be a form of “prior art,” even though it described the very engineering work for which Eckert and Mauchly sought patent protection. Ironically, von Neumann’s writings became so influential that the concept of a stored program binary computer, as created by Eckert and Mauchly, is now commonly referred to as a “von Neumann architecture.”

Based on that, and Atanasoff’s predecessor prototype, the judge ultimately ruled in Honeywell’s favor. The patent was invalidated, and so any of Honeywell’s infringement incurred no damages. The judge found that Eckert and Mauchly should share the credit for inventing the computer with Atanasoff.

The decision killed Sperry Rand’s monopoly. The public benefits were and remain immense. The most important invention of the 20th century was open to the public, and anyone could make and sell a computer. 

The trial also gave to posterity a fulsome history of the invention of the computer. Whatever fame and accolades should have been due to the inventors, however, never properly came. The outcome of the trial was overshadowed in the press by the Watergate scandal, with the publication of the ruling on October 19, 1973, competing with the “Saturday Night Massacre” for public attention. Even today, the name von Neumann comes more easily to mind in connection with the computer’s invention than the names Eckert, Mauchly and Atanasoff.