Zum Inhalt springen

The Weather Machine: What ENIAC Operators Were Told

Zusammenfassung

By accounts from the project, Army recruits assigned to operate the ENIAC (Electronic Numerical Integrator and Computer) at the Moore School of Electrical Engineering in Philadelphia were told that the machine was a “weather predictor” — an anecdote that is widely repeated but not well documented in primary sources. The intent was to keep the machine’s actual purpose — calculating ballistic trajectories and artillery firing tables for the Army’s Ordnance Department — from becoming public knowledge during and immediately after World War II. The ENIAC was the world’s first general-purpose electronic digital computer, operational in 1945, and the recruits who changed its vacuum tubes and patched cables were not told what they were actually computing.

The Machine and Its Purpose

The ENIAC was built at the University of Pennsylvania’s Moore School of Electrical Engineering between 1943 and 1945, funded by the US Army’s Ballistic Research Laboratory. The explicit purpose was to compute artillery firing tables — the tables used by artillery crews to aim their guns at different targets under different conditions of elevation, wind, temperature, and propellant charge.

Computing these tables manually took human “computers” (women with mathematics degrees employed for the task) approximately 12 hours per trajectory. The Army needed thousands of trajectories per table. The ENIAC could compute a trajectory in approximately 30 seconds.

The machine was completed in late 1945 — after the war in both Europe and the Pacific had already ended, too late to compute the firing tables it had been commissioned for. Its first official computation, in December 1945, was calculations for the hydrogen bomb design — supplied by John von Neumann, who was consulting on both the ENIAC project and Los Alamos.

The Cover Story

By the project’s own later accounts, the Army recruits assigned to keep the ENIAC operational — replacing failing vacuum tubes, resetting switches after programming changes, monitoring for hardware faults — were young enlisted men with no particular technical background, and they were told the machine computed weather patterns. The anecdote is repeated in popular histories but is not firmly attested in contemporary documents.

This was not an elaborate deception. The recruits were not security threats; the Army simply preferred not to explain ballistic computation to every person who handled the machine. “Weather predictor” was a plausible, non-sensitive description for a very large computing machine during a period when the machine’s actual applications were sensitive.

The Six Women Programmers

The ENIAC’s actual programming was performed by six women — Jean Jennings (Bartik), Betty Snyder (Holberton), Marlyn Wescoff (Meltzer), Frances Bilas (Spence), Ruth Lichterman (Teitelbaum), and Kathleen McNulty (Antonelli) — who were not given the cover story. They were mathematicians hired by the Army and told exactly what the machine did and what they were computing.

These women — called “computers” in the job listings that recruited them — taught themselves the ENIAC’s architecture from wiring diagrams and logic diagrams. There was no programming language; programming the ENIAC meant physically setting switches and plugging cables to route computation through the machine’s accumulator units. They developed the first programming techniques for the machine, including subroutines and nested loops, without any formal computer science education (which didn’t exist yet).

The women programmers were not acknowledged at the ENIAC’s public demonstration in February 1946. They stood in the back of the auditorium. Some accounts record that guests asked if they were “models” — meaning fashion models hired to pose with the machine. Their contributions were not publicly recognized for decades. Jean Jennings Bartik received an honorary doctorate from her alma mater Northwest Missouri State University in 2002.

📚 Sources