The digital computer is usually credited as the invention of two professors at the University of Pennsylvania, J. Presper Eckert and John Mauchly. Funded by the United States Army, the ENIAC computer was designed to calculate tables for launching artillery shells accurately in World War II, but was not completed until after the war in 1946. Unlike earlier computers that had a fixed purpose, ENIAC (meaning “Electronic Numerical Integrator And Computer”) could be reprogrammed to handle many different purposes. But were Eckert and Mauchly really the pioneers of today’s modern digital age?
Actually no. The real inventors of the digital computer were physics professor John Atanasoff and his student Clifford Berry who created the first digital computer in a laboratory at Iowa State College. The ABC (“Atanasoff-Berry Computer“) was built in 1939, yet by the time of ENIAC’s introduction to the world, the ABC had been forgotten. What had happened? World War II broke out and Iowa State as well as Atanasoff and Berry simply didn’t realize the power of what they had created. Atanasoff was called up by the Navy to do physics research, eventually participating in the atomic bomb tests at Bikini Atoll.
When Atanasoff returned to Iowa State he found that his invention was gone to make room for other equipment—because the ABC was built piece-by-piece in the laboratory, it was too big to move out and so it had to be dismantled. Iowa State had decided that a patent was too expensive and so never filed one. John Atanasoff went on to gain recognition for a number of inventions involving physics, but the ABC was mostly forgotten.
In the 1970s there were a handful of companies that saw the great potential in the electronic computer. Sperry Rand Corporation, which was formed through a series of mergers and acquisitions including the Eckert–Mauchly Computer Corporation, held U.S. Patent 3,120,606 for the digital computer. In 1973, Sperry Rand sued Honeywell, Inc. and Honeywell reciprocated. Thus began one of the most important intellectual property cases in history.
During the research for this case, Honeywell found out about John Atanasoff and the ABC, which became pivotal information. The case was tried for 7 months after which Judge Earl R. Larson handed down his decision that stated, among other things, that the Eckert-Mauchly patent was invalid.
Some people have disputed this finding, arguing that this was a “legal” finding or a “loophole” or that a lawyer or a judge simply couldn’t understand the complex engineering issues involved. Here’s my take on this.
- Both sides had a lot of time, and access to technical experts, to make the best case they could.
- So much was at stake, and a huge amount of money was spent to bring out the truth. Both sides had very significant resources. If a case with this much at stake could not convince a judge after seven months, then there is little hope for any IP case.
- Evidence was found and witnesses verified that John Atanasoff had attended a conference in Philadelphia where he met John Mauchly and described his work. He then invited Mauchly out to Iowa where Mauchly spent several days examining Atanasoff’s computer and many late nights reading Atanasoff’s technical specifications. Letters were produced, signed by Mauchly, that thanked Atanasoff for his hospitality and for the tour of his amazing invention.
- Mauchly testified at the trial. He admitted that he had met Atanasoff and eventually admitted that he had examined the ABC and read its specification.
- Mauchly and Sperry Rand Corporation were challenged to produce a single piece of evidence that Mauchly or Eckert had written about or researched digital electronics before Mauchly’s meeting with Atanasoff. The best Mauchly could do was produce a circuit for a model railway flasher that he claimed was a binary counter—it counted from 0 to 1 and then back to 0.
In fact, it became clear that Mauchly and Eckert attempted to claim much more credit than they deserved and tried to deny credit to others. They had actually greatly improved on Atanasoff’s original design. Had Eckert and Mauchly been more humble, had they added Atanasoff’s name to their patent, had they patented their own improvements instead of the entire invention, they may have given Sperry Rand the most powerful IP in technology history. Instead the invention of the computer entered the public domain without restriction, and the rest is history…
For a good book on the subject, read The First Electronic Computer: The Atanasoff Story by Alice R. Burks and Arthur W. Burks.
Atanasoff supporters always leave out the fact that he never completed his machine; what Mauchly saw was a small piece of it. The ABC was not truly electronic in that it did not utilize electronic speeds, relying on human intervention every step of the way. The ENIAC was a digital machine, not a binary machine like the ABC, and could hardly be said to be any kind of descendent of the ABC. Finally, the ABC was FIRST COMPLETED in the 1970s, when Honeywell paid to have it constructed for the purpose of the trial. Art and Alice Burks seem to carry some kind of grudge against Mauchly and Eckert, possibly stemming from not getting all the credit they sought for their role in construction of the ENIAC. For a better researched look at how ENIAC came to be, read Scott McCartney’s book.
Thanks for writing. Obviously there is a lot of passion among people on this topic. However, your facts are wrong. The ABC was up and running, but was dismantled to make space for other equipment in the lab. It was fully electronic, at least as much as ENIAC that required switchboard plugs to program it. Finally, I think that if the things you say were true (and I’ve heard them all before), wouldn’t Sperry have made that case in court? Or Mauchly? Why did Mauchly lie about his contact with Atanasoff until he was presented with his own letters to Atanasoff in court? Anyway, all of this is documented in several books and, most importantly, the trial transcripts. You can find further information, and confirmation, at the University of Pennsylvania archive, http://www.archives.upenn.edu/faids/upd/eniactrial/eniac.html, as well as many other places on the web. You’ll also find that there were a lot of people who were pretty upset with Eckert and Mauchly for not giving them credit, including several who sued to get their names on the computer patent.
There were powerful elements in the US, of which the Judge is only one, who were determined to subvert Mauchly’s patent. Pilfering Mauchly’s ideas began early with the likes of Von Neumann.
Atanasoff is an untrustworthy man. The most important question obviously is this: Atanasoff was determined to patent his device, which is abundantly clear. Why then would he disclose everything to Mauchly as he claims, explaining “everything in detail and answering all his(Mauchly’s) questions”. Ridiculous.
The efforts to show that the ABC worked are suspiciously strenuous and perfect. It’s artificial.
Another important point is that Atanasoff did not understand the ABC because Berry designed and built it.
Mauchly had a very clear vision of what was technically required and how to implement it. He created an entire industry and foresaw the PC revolution. To attribute this to Atanasoff is psychedelic.
Bob, if you are aware of the U Penn archives, then you should be aware of the wealth of information on Mauchly’s pursuits of electronic computation before ever meeting Atanasoff, going back to the mid-1930s. Most of this was NOT presented at the trial, because Sperry lawyers underestimated its value to the case. The trial was all about getting money for Honeywell and CDC, not about truth on any level. If you were to examine the actual records, and not just those that were included in the 1973 trial, you would come to a different conclusion. Larson’s verdict was not appealed because by 1973 it would not have been worth the additional expense to Sperry to have it overturned. This does not mean the findings were anything like “the truth.”
Very simply: if it is not programmable, it’s not a computer.
Herbert, ENIAC was also hardwired to do calculations. By your definition, the EDSAC in Cambridge was the first computer in 1949. However, that’s not the definition that computer scientists or historians use for “computer,” but rather that’s the definition of “stored program computer.”
Re: “. If a case with this much at stake could not convince a judge after seven months, then there is little hope for any IP case.”
You’re right … there is little hope for any IP case. I learned that some years ago, as a an expert witness in a number of cases, both in the U.S.A. and Canada. Part of the reason is that nothing in real life is “binary”.
I have learn some excellent stuff here. Definitely value bookmarking for revisiting.
I surprise how a lot effort you place to create this type
of great informative website.