Frank Bernheisel: The View From Here
Frank Bernheisel
Frank Bernheisel
Posted 02.28.15
Just Outside Washington


Alan Turing, and early computing

In 1960 I graduated from George Washington University with a degree in Mathematical Statistics and went to work for the UNIVAC Division of Sperry Rand. I was immediately sent to classes to learn to program the company's scientific computers, Univac 1103 and Univac 1105. We learned binary machine language and assembly language to program these machines. We also learned the history of the programmable digital computers. (For a detailed history, see George Dyson's Turing's Cathedral: The Origins of the Digital Universe.

Our course introduced us to Alan Turing, famous lately for the movie, The Imitation Game. We went to see it the other day; it was an excellent movie starring Benedict Cumberbatch. Highly recommended.

However, the movie does not provide a good history of computing or of Turing's contribution. The movie is about the breaking of the Enigma machine codes, which was done with a big electromechanical calculator.

Turing's initial and primary contribution was theoretical: proving that a programmable machine using algorithms could solve problems. This he spelled out in his 1936 paper: On Computable Numbers, with an Application to the Entscheidungsproblem. Turing received his PhD. from Princeton in 1938 and visited the U.S. several times during WWII, neither mentioned in the movie.

In 1935 John von Neumann was invited to the Institute for Advanced Study (IAS), in Princeton and worked on many projects. At the end of WWII the IAS computer, was built under his direction and using a design from one of his papers. Also, Von Neumann used some of Turing's ideas in his paper on the EDVAC computer. Turing worked on the design of the ACE computer in England after WWII.

At the Moore School of Electrical Engineering at the University of Pennsylvania, just 45 miles from Princeton, J. Presper Eckert and John Mauchly constructed the ENIAC computer for the Army in 1943. They followed up with a second computer three years later, the EDVAC. John von Neumann consulted with them on the project and published a report on the EDVAC. This report helped nullify the patent claims of Eckert and Mauchly in their fight with the Moore School.

Eckert and Mauchly left the Moore School and started their own company because of the dispute over patents. They designed a computer called the BINAC, which was followed by the development of the UNIVAC computer. The Bureau of the Census sponsored and paid for much of the development of the UNIVAC. It was put into commercial production and sold as UNIVAC I. In 1950 Eckert - Mauchly group was purchased by Sperry Rand. The Bureau of Census had purchased several of the computers which were in operation when I visited the Bureau as a UNIVAC employee in 1960.

Also, during WWII, the U.S. Navy had people working on code breaking. In 1946 they formed ERA in Minneapolis - St. Paul, developed several computers and perfected drum memory. ERA had a contract to develop a computer for the Navy to break the Soviet's codes and changed their designs after some input from the Moore School. The resulting computer became the ERA 1101 (binary for 13). ERA had financial problems and was bought by Sperry Rand in 1952.

Sperry Rand now had two separate groups working on computers; the Eckert – Mauchly group marketing business computers, the UNIVAC 1 and the ERA group marketing scientific computers, the UNIVAC 1103. The 1103 was the successor to the 1101. Later, some of the ERA group split off from UNIVAC and formed Control Data Corporation and Cray Computers. During this time, IBM was doing something with computers, but we at UNIVAC did not pay them much mind. The UNIVAC scientific machines were superior, and IBM was just a bunch of Marketeers.