Re: Scripting vs. Programming language vs. 4GL?

|> prechelt@ira.uka.de (Lutz Prechelt) writes:
|> : I would say that every language which is (almost?) ALWAYS interpreted
|> : instead of compiled can be called a scripting language.
|> :
|> : And the borderlines are also not sharp:
|> : Think of Common Lisp which nobody would call a scripting languages,
|> : although it is at most partially compiled.


lwall@netlabs.com (Larry Wall) writes:
|> True.


No. False.


I would bet that 99.99% of all Common Lisp code runs compiled.
Most people run code interpreted only while debugging, if then,
except for small top-level scripts that interface to the main code.
One of the advantages of lisp as a programming environment is that
you can run intermixed compiled and interpreted code--each function
can be either at runtime, and you can dynamically compile or uncompile
routines at runtime. (E.g. the compilation status of a routine can be
altered for debugging purposes.)


In fact, lisp is a wonderful shell/scripting language. It's pretty common
to start a lisp session and use it for several days, much as you would a
unix shell. Loading little lisp files to perform tasks similar to "make"
is quite common.


Also note that the startup scripts for emacs are written in elisp.


People who have been spoiled by lisp environments find the restrictions of
unix shells maddening--there's even a unix-haters mailing list where they
share their tales of woe. (Or at least there was up to a year or so ago--
by accident I'm not currently subscribed.) I suppose not everyone on that
list has used lisp, but the correlation seems pretty strong. The most
subtle horror stories relate situations where typical unix users can't
even see the problem being encountered--the lossage looks proper and
normal to them.
--
James McDonald
Kestrel Institute mcdonald@kestrel.edu
3260 Hillview Ave. (415) 493-6871 ext. 339
Palo Alto, CA 94304 fax: (415) 424-1807


--

Return to the comp.compilers page.
Search the comp.compilers archives again.