GARBAGE IN, GARBAGE OUT
Q From Martin Schell: What is the origin of the cynical techspeak garbage in, garbage out? Is it a play on the old acronyms LIFO (last in, first out) and FIFO (first in, first out)? It seems to be somewhat widely known as a general principle in the information sciences: if you input data that's junk, the result is unreliable junk.
A That’s certainly what the expression means: that the results of a computation are only as reliable as the data you input.
Both the expression and its abbreviation GIGO have been around, I suspect, since the early days of commercial computing. But the first example I can find is in a syndicated article about the first stages of computerisation of the US Internal Revenue Service that appeared in several US newspapers on 1 April 1963. As a reminder of the state of the art in those days, it rather sweetly referred to the computer at the centre of the system as a robot and described data being input in local offices on punched cards, transferred to magnetic tape, and flown to the computer centre for processing. The article quoted the full expression, which it said even then was “an adage of computer men”. The abbreviated form GIGO is recorded the following year and both the expression and the acronym become quite widely known soon after.
Though it was popular with computer people because it encapsulated a fundamental truth about data handling, it seems also to have caught the imagination of the general public — worried about the implications of automation for their jobs — because it reassured them that these new-fangled computers were just dumb beasts that needed handling by intelligent beings.
As you might guess, we’ve no idea who coined it. The OED’s first example, from 1964, appeared in a book entitled The Impact of Computers on Accounting. That supports your view that GIGO was based on the two accounting terms LIFO (“Last In, First Out”) and FIFO (“First In, First Out”), which became computing terms for ways to manage the stacking and order of processing of data. Accountants had been using them to describe ways of managing stock levels or valuing a company’s goods at least since the 1930s, though the acronyms only appeared in print for the first time around 1945.