The father of Fortran changed programming forever
For years, John Backus seemed an unlikely candidate to become a pioneer in any field, much less in advanced computing. As a teenager, he attended an exclusive private high school in Pennsylvania where he spent most of his time finding new rules to break. He moved on to the University of Virginia to study chemistry — but was expelled for poor attendance. He later enrolled in medical school only to drop out after nine months, complaining that medical training focused too much on memorizing concepts.
Eventually, Backus found his true calling in computing, and he is now best known as the father of Fortran, the first widely used, high-level programming language that helped open the door to modern computing. During his 41-year career at IBM, Backus also helped develop the IBM 704, the first mass-produced computer with core memory and floating-point arithmetic hardware, as well as Speedcode, the first high-level programming language created for an IBM computer, and the Backus-Naur Form (BNF), a pioneering notation for describing the structure of programming languages.
“Much of my work has come from being lazy,” Backus once said, with characteristic modesty. “I didn’t like writing programs, so I started work on a system to make them easier to write.”
The primary author of Fortran, John Backus
Backus began his career at IBM purely by happenstance. In 1950, just after graduating with a master’s degree in mathematics from Columbia University, he wandered into IBM’s headquarters on New York’s Madison Avenue, where the Selective Sequence Electronic Calculator (SSEC) — a room-sized electromechanical computer — was on display. Backus mentioned his interest in mathematics to a tour guide and was promptly given an informal oral exam consisting of math brainteasers. He aced the test and was hired on the spot as a programmer, even though he had little idea what programming was.
At IBM, Backus readily applied talents that weren’t always apparent during his academic career — an extraordinary gift for problem-solving, including finding ways to break down those problems into more manageable forms and, above all, a willingness to fail in the pursuit of knowledge. In 1953, frustrated by the complexity and high cost of programming, he assembled a small team at IBM to come up with a better way.
The eclectic group included a chess wizard, a crystallographer, a cryptographer, a researcher from the Massachusetts Institute of Technology, and the only woman on the team, Lois Haibt, who joined straight out of Vassar College and would go on to become an early pioneer in computer science.
The team tackled two fundamental problems: how to make programming faster, cheaper and more accessible to a wider range of users, and how to structure the underlying code to make all that possible. Backus managed with a light hand — he deemed IBM’s rigid yearly performance reviews ill-suited for his team, so he simply ignored the reviews process.
In 1957, the IBM Mathematical Formula Translating System, or Fortran, debuted. It fundamentally changed the terms of communication between humans and computers. What was formerly a laborious task of manually keying in as many as a thousand program instructions for a given problem could now be translated, automated and reduced to only 47 in Fortran.
Fortran quickly became central to addressing a wide range of data-laden problems, from calculating trajectories of airborne missiles and NASA flight patterns to computing complex economic and statistical models. More than six decades after its debut, Fortran is still used in Doppler radar weather forecasts and atmospheric and oceanic studies, as well as in simulating nanoparticles, genomes, DNA and atomic structures.
After his work on Fortran, Backus, along with Danish computer scientist Peter Naur, developed the Backus-Naur Form, a mathematical notation for describing the structure of programming languages. It formally defines the grammar of a language, so that there is no ambiguity as to what is allowed and what is not.
Entrance to the Fortran 25th Anniversary exhibit
A 1957 manual, Programmers Primer for FORTRAN Automatic Coding System for the IBM 704
Backus also was a pioneer in functional programming (FP), a system of programming that focuses more on describing the problem a person wants the computer to solve and less on giving the computer step-by-step instructions. FP sought to overcome the limitations of the word-at-a-time style of conventional programming by combining forms for creating programs to build still higher-level ones. Languages like Java and Python are increasingly adopting concepts derived from functional programming, and newer languages like Haskell are going completely functional.
From 1963 through his retirement from IBM in 1991, Backus was an IBM Fellow at the research facilities in Yorktown Heights, New York, and San Jose, California. He received the National Medal of Science for pioneering contributions to computer programming languages, the first IBM employee to win the award. He also received the prestigious ACM Turing Award, and even had an asteroid named in his honor: Asteroid 6830 Johnbackus.
Backus died in 2007, at age 82. He left a legacy not only as a pioneer in making programming faster and more accessible, but also as a researcher fearlessly committed to the process of trial and error.
John Backus standing beside the IBM 704
The world’s first programming language standard opened the door to modern computing
Establishing IBM’s reputation in the mainframe market — and the electronics industry at large
IBM has always provided its scientists the freedom to explore and discover solutions to society’s greatest problems