JC>  SI> Compilers are written in response to language definition. Not t
JC>  SI> other way around. If compilers defined languages, we wouldn't n
JC>  SI> BNF grammars. :)
JC> Playing my usual position of devil's advocate: the original definiti
JC> of LISP was a LISP interpreter, which was itself written in LISP.  T
JC> legal syntax was anything it would accept, and the definition of the
JC> semantics was what it did.
No compiler or interpreter is possible to be written at the basest level 
within the same language framework, without a basic engine written in a 
pre-existing language. Even if it was done in assembly first, it was 
done in something else before it was done in the 'language'. C is 
compiled in C. C++ can be compiled in C++. Or in C. I'm sure LISP didn't 
just exist. Therefore it stands to reason that LISP, although can be 
used to write LISP interpreters, was not done that way the first time.
JC> Nicely enough, LISP is quite an easy language to interpret, and also
JC> lends itself well to writing interpreters, so the "definition" was l
JC> than one page long...
I'd be interested in getting that definition. Do you have it lying 
around somewhere?
 * RM 1.3 02587 * RoboMail -- Version 1.3 -- Available now!
--- Maximus/2 3.01
---------------