The History Behind the Definition of a 'String'

Viewed 8711

I have never thought about until recently, but I'm not sure why we call strings strings. I am a .NET programmer, but I believe the concept of strings exist in virtually every programming language.

Outside of programming, I don't believe I've heard the word string used to describe words or letters. A quick Google of, 'Define: string' yields a bunch of definitions that have nothing to do with the concept of letters, words, or anything of the nature associated to programming.

My guess of it, is that, back in the day, strings were really just arrays of characters of a particular length, often with a delimiting character at the end. But, I don't see a natural transition from 'character array' to string.

Can someone offer up some insight to why we call strings strings?

10 Answers

My assumption has always been that the programming term originated from the following definition of the word "string" (from Merriam-Webster):

(1): a series of things arranged in or as if in a line <a string of cars> <a string of names>

(2): a sequence of like items (as bits, characters, or words)

Since a string in programming is simply an ordered sequence of characters, referring to this as a "string of characters" (or simply "string") seems like the most probable origin.

From this reference:

The 1971 OED (p. 3097) quotes an 1891 Century Dictionary on a source in the Milwaukee Sentinel of 11 Jan. 1898 (section 3, p. 1) to the effect that this is a compositor's term. Printers would paste up the text that they had generated in a long strip of characters. (Presumably, they were paid by the foot, not by the word!) The quote says that it was not unusual for compositors to create more than 1500 (characters?) per hour.

From searching through the ACM bibliography it seems the word string acquired its meaning in computer science during the 1960s. At the beginning a string is a general kind of sequence or list, e.g. A command language for handling strings of symbols from 1958.

This article explicitly mentions "character strings" in 1964.

Unfortunately I can't access the full texts, which are behind a toll booth.

I suspect it's because string originally meant just a sequence of data values: "I'll just string these together" etc. These values didn't have to be characters. One very common use for this general concept happened to be a sequence of characters, and this took over as the general meaning of the word.

The word was originally used to differentiate between a set of values to which the particular order of elements doesn't matter (for instance, a set of random samples of measurements) and another that could only have its meaning preserved when the order is also preserved. Originally a string could be a set of any kind of values, but since in the post-mainframe era a string of characters is by far the most common kind, the fact that the values are characters became a "default".

It's called a strings, because it's actually an array of char type elements.

That being said, they are "stringing together" (or is it strung together) via this array, which turns them into a "string".

Related