> In fact, I've wondered about the origins of the term "intern" since I first heard of it
I always assumed it came from internalize, or something like that.
I always assumed it came from internalize, or something like that.
I'll take a guess, though, that it's from Common Lisp, because CL's intern function (http://clhs.lisp.se/Body/f_intern.htm) is clearly part of a suite of names (intern, unintern, :internal, :external) where, in context, "intern" makes complete sense as the name for the function.
http://www.softwarepreservation.org/projects/LISP/MIT/Moon-M...
See sections 6.3 and 6.4 and the glossary for definitions and uses of the word.
Unfortunately they don't shed any light on where the term "intern" comes from. Typical usage is that a string is «"interned" or registered in an obarray» but it doesn't explain much more than that. Interning occurs most often at read time, when a string read from the input is turned into an atom, so I'd guess that "intern" is short for "internalize" or similar.
In CL, you don't really have a global "symbol table" or "atom table"; instead, interning is about the runtime lexical scopes under which code compilation happens, which consist of stacks of imported namespaces ("packages"), where each namespace exists as a mutable object holding identifier "slots".
(intern foo) is a mutation on the current namespace that creates a local variable "slot" for the identifier foo, shadowing any other foo that might have been in lexical scope from an import. That is, (intern foo) makes foo resolve internally to the current namespace.
(unintern foo) is the complementary mutation; it drops the foo identifier from the current namespace, allowing other symbols (slots in packages) that were previously shadowed by it to re-appear.