Card reader

This IBM card reader has a vertical hopper on the upper right where the cards to be read were placed. When the computer was ready to read in another job the card reader would suddenly read through and scan a deck of cards, at a rate of several cards per second. Once read the cards would be deposited in an output hopper, where the person who fed the cards in, would retrieve them, because there was likely an error in the deck that they would have to fix, before their job was successful.

A card reader is an electronic device that reads data written on card stock. The US Census made very early use of cards, with holes punched through them, that were processed with purely mechanical devices. International Business Machines first products were electromechanical devices that could read cards, and do very simple mathematical operations on them.

When Richard Feynman was very young he played a role in the Manhattan Project, one of the tasks assigned to him was to set up workshops full of electromechanical devices that used cards, that he was to use to solve very time consuming calculations. Each machine would do a single operation, possibly a calculation, on the data on a card, and possibly add one further digit. Very large decks of cards had to be fed through these machines, in a particular order.

His workshop was manned by dozens of very junior GIs, who were selected due to their high IQs, right out of basic training. Feynman described how a single calculation could take his platoon of young genuises weeks or months -- calculations that a modern computer could be programmed to perform in a fraction of a second.

Input to early computers was not done through interactive monitors, when computers were operated in batch mode. Two common input mechanisms were punched cards, or ticker tape.