Gustav Tauschek: The Forgotten Inventor
Abstract
Gustav Tauschek (1899–1945) was a Viennese inventor with no formal engineering training who nevertheless produced one of the densest patent portfolios of early information technology, including the magnetic drum memory (1932), the storage technology that would serve as main memory for the first generation of commercial computers, and one of the earliest optical character recognition machines. IBM bought his patents by the dozen. Tauschek shows how much the early IT giants owed to individual outsiders whose names the industry then forgot.
The Autodidact
Tauschek was born in Vienna on April 29, 1899. He never completed a formal engineering education; his qualification was an obsessive talent for electromechanics, applied from 1922 onward to the dominant information technology of the day, the punched card machine (see Herman Hollerith and the Punched Card).
From 1926 to 1930 he worked for the German arms and machinery firm Rheinmetall in Sömmerda, where he developed a complete punched-card accounting system of his own design (carriage, printer, arithmetic) a prototype of which survives today in Vienna’s Technical Museum. It was never mass-produced: in 1928 Rheinmetall’s punched-card subsidiary was acquired by IBM, which thereby acquired Tauschek too. He received a five-year contract as an inventor-consultant, effectively, IBM was buying a one-man research laboratory.
The Drum: A Memory Before the Computers Existed
Tauschek’s most consequential invention came in 1932: the magnetic drum memory, a rotating cylinder coated with ferromagnetic material, written and read by fixed heads. His original design stored on the order of 500,000 bits. There were as yet no electronic computers to use it; Tauschek conceived it as storage for electromechanical accounting machinery.
The idea outlived its context spectacularly. When electronic computers arrived after the war, they had fast processors and no adequate working memory, and the magnetic drum became the standard solution: the IBM 650 (1954), the most widely sold computer of the 1950s, was built around a drum, as were dozens of other machines (see The Storage Revolution and, for what drum-based programming did to programmers, The Story of Mel). For roughly a decade, until magnetic core memory displaced it, the working memory of the world’s computers was Tauschek’s rotating cylinder.
Reading Machines and 169 Patents
In 1928–1929 Tauschek also designed a photoelectric reading machine, a device in which a photocell scanned printed characters and matched them mechanically against templates. It is recognized today as one of the first optical character recognition (OCR) machines, patented decades before OCR became a commercial field. Among his other designs was an accounting machine capable of storing 10,000 bank account records on magnetized steel plates.
In total, Tauschek sold 169 patents to IBM over his career. The arrangement was characteristic of IBM’s early strategy under Thomas J. Watson Sr.: rather than rely solely on in-house development, the company systematically bought the patents of independent inventors, building the intellectual-property wall behind which its tabulating-machine dominance (and later its computer dominance) grew (see IBM: The Company). Tauschek was among the most productive single sources of that wall.
Dead End: The Inventor Without an Industry
Tauschek died on February 14, 1945, in a Zurich hospital, of an embolism, a few months before the war’s end, and a year before ENIAC made electronic computing public. He never saw a computer, never saw his drum become the industry’s standard memory, and never became a public figure of computing history.
His obscurity has structural causes worth naming. He was an inventor, not a founder: his patents made IBM’s products better and IBM’s shareholders richer, but nothing carried his name. He worked in Austria and Germany in the interwar period, outside the Anglo-American institutional network that later wrote computing’s history. And he died before the field existed that would have claimed him retroactively. The pattern repeats across early computing (the individual tinkerer absorbed by the corporation) and Tauschek is its purest Austrian case: a man whose inventions were everywhere in the machine room of the 1950s, and whose name was nowhere.