| |
Www.Wired.Com - 1944: Harvard and IBM dedicate the Mark I computer. Also known as the IBM Automatic Sequence Controlled Calculator, or ASCC, the pioneering computer was notable for producing reliable results and its ability to run 24/7.
Harvard electrical engineer Howard Aiken first dreamt up a large-scale calculator in 1937. He knew he needed a corporate partner and first courted Monroe Calculator Company, which turned him down. Aiken went back to the drawing board and came up with a proposal that convinced IBM, whose big product at the time was a punch-card processor. A big plus in the proposal was that it used so many existing IBM components in a new way.
Clair Lake, Frank Hamilton and Benjamin Durfee finished the Harvard computer at Endicott, New York, in January 1943. They demonstrated it to the Harvard faculty members in December, and then took it apart, packed it up and shipped it off to Cambridge, where it was rebuilt in the basement of the physics lab.
The Mark I was a monster: 55 feet long and 8 feet high. It weighed five tons and contained 760,000 components, including 3,000 rotating counter wheels and 1,400 rotary-dial switches, along with an assortment of shafts, clutches and electromagnetic relays, all linked together with 500 miles of wire. Its clickety-clack sounded like a "roomful of ladies knitting."
You fed instructions in on paper tape, and loaded the data on punch cards. It could only perform operations in the precise linear order it received instructions. The tape could not run backward.
The Mark I could handle 23-decimal-place numbers and perform addition, subtraction, multiplication and division. It was also programmed with subroutines for logarithms and trigonometry.
It was slow, taking three to five seconds to do a multiplication. It gave you results through two outputs: teletypewriter and punch card.
Mathematician Grace Hopper of the U.S. Naval Reserve joined Aiken's team at Harvard and was instrumental in keeping the Mark I running. She repaired it one day by removing a moth that had fouled the Mark I's electromechanical innards, becoming the first person to debug a computer. She then coined the term computer bug.
When the time neared to dedicate the Mark I, in August 1944, the Harvard News Office put out a press release giving all the credit for the machine to Aiken. IBM chief Thomas J. Watson was himself so put out that his firm's work was not being acknowledged that he threatened to return to New York, boycotting the dedication and luncheon festivities. Cooler heads prevailed, and Watson stayed for the hoopla, but Aiken and Watson never got over their turf tiff. Years later, when Thomas J. Watson Jr. made a peace offering of a consultant gig at IBM, Aiken refused to sign a nondisclosure agreement.
Hopper and Aiken (also USNR) used the Mark I to help the Navy produce tables for aiming artillery shells and bombs in the closing year of World War II. The electromagnetic machine remained in use until 1959, by which time it was left in the dust by true electronic computers using first vacuum tubes, then transistors, then chips.
And for all of the Mark I's advances, German engineer Konrad Zuse's Z3 model from 1941 may have preceded it as the world's first fully functional, programmable computer.
Aiken went on to build the Mark II in 1947, the same year he founded the Harvard Computation Laboratory and predicted, "Only six electronic digital computers would be required to satisfy the computing needs of the entire United States."
Source: Various
Get the latest in science news, including space, physics, planet earth, discoveries, NASA, satellites, and space travel from Wired.com {...} Published: August 7, 2008, 5:00 am - Indexed: August 8, 2008, 11:27 am - Page Size: 49KB |