moderately stiff paper approximately 3 ¼ inches by 7 3/8 inches. The card presented 80 columns of 12 rows each. One or more positions in a column could be punched, leaving holes in the card that represented characters. Such cards could be used to enter programs, that is, instructions defining the operations that the computer was to perform, or data, that is, the information that the operations were to use in their processing. The communication protocol between the person and the computer was, by today’s standards, cumbersome and non-intuitive. This is a bit more understandable given the fact that punched card use in computer technology derived from their use in the Jacquard looms of the early 1800’s. Computers tended to be viewed as extensions of mechanical devices, so this form of input fit well with the paradigm.

Output from mainframe computers, that is the output that was used directly by people, tended to be either printed documentation, produced by a line printer device, or graphic displays produced by pen-plotter devices. Interactions between a person and a computer tended to involve a cyclic protocol. Instructions for processing operations were punched into cards, as was input data that was to be the fodder for those operations. The computer read, interpreted and acted upon the instructions and data from the punched cards (a rather graphic illustration of the term number crunching) and then printed the results of the processing to a line printer or pen-plotter. Variants on this protocol might entail the creation of intermediate result data on other cards punched by the computer, or the storage of data on magnetic tape or early disk drive units, and then the protocol would start anew.

The first family of truly high production volume mainframes was the IBM 360 series. These started making their appearance in corporations in the 1960’s before subsequently maturing in the 1970’s in the form of the IBM 370. Total production of these machines ran into the tens of thousands around the world to perform business tasks of all sorts. The cost of those computers and the environment they required ran in the tens of millions of dollars.

IBM started as a small company that emerged at the time of mainframe developments as the undisputed leader of a pack of companies whose names included Burroughs, Control Data, General Electric, Honeywell, RCA, Sperry and Univac.

It is historically interesting to remember that the United States space endeavors that first placed a man on the moon in 1969, Project Mercury, Project Gemini and finally Project Apollo, were all products of the computer mainframe epoch. But, to put things in scale, the entire computational capability of the National Aeronautics and Space Administration (NASA) in 1969, including processing power, main computer memory and on-line disk storage, was probably less than what is today available on one fully configured personal computer system. More to the point, certainly a very significant, if not the primary contributor, of computing resources came in the form of bamboo and plastic; that is, the slide rules that hung from the belts of most NASA engineers.

The Semiconductor Mutation

The mutational event that truly launched the age of the computer was the 1948 invention by William Bradford Schockley (with John Bardeen and William H. Brittain) of the transistor. The transistor is a device that makes use of the properties of a class of materials known as semiconductors. These characteristics offer very different operating mechanisms than vacuum tubes, but they can be used to effect much the same capabilities as various vacuum tube configurations. Through these configurations, the operations necessary to construct stored-program computers were possible in much smaller physical configurations than those constructed

Book available at Midori Press (regular)
Book available at Midori Press (signed)
Book available at Amazon (regular)