Subject: Re: EVAL Implementations
From: Erik Naggum <erik@naggum.no>
Date: 1996/08/23
Newsgroups: comp.lang.lisp
Message-ID: <3049754333170201@arcana.naggum.no>


Robert Munyer's exceedingly lengthy diatribe seems to hinge on one issue:
whether the semantics of `setq' at top-level of a variable that has not
been declared special is defined by Common Lisp.  I can't see that it is.
I'd like Robert to show me how he found the semantics of this to be defined
by the language specification since he makes so many conclusion based on
his assumption that its semantics _is_ defined.

barring any overlooked clauses which Robert will bring to my attention,
whatever it is that he's trying to distinguish between `setq' and `defvar'
at top-level is purely a matter of how _implementations_ differ in
implementing undefined semantics.

I'm sorry to see that Robert incessantly perceives any pointers to a
specification that does not support his claims, together with a refusal to
speak on behalf of or for those who defined the standard, as "vagueness".
I do not find it to be my job to over-interpret a standard in the face of
unwillingness, or even lack of interest, to read it.  If a standard cannot
speak for itself, it should be improved until it does, but it is hopeless
to attempt to let things speak for themselves if the audience does not
share its context.  I specifically do not claim that I understand the
totality of Common Lisp, which is why I point to authoritative sources that
I find to contradict non-authoritative examples that do not appears to show
an interest in the language, but rather in implementations, which concern
me only inso far as they implement the specification.  otherwise, they have
bugs that should be reported.  if legitimate disagreement over some clause
in the standard is found, the right thing is to submit a request for
clarification to ANSI about it -- there are procedures for this, and the
committee should be prepared to handle such requests.  a prerequisite to
all this is of course that all parties involved in the disagreement know
precisely what the standard says on the issue (or not, as the case may be).

finally, this is not a question of whether _I_ agree or disagree with
anything Robert says or implies.  the question is: is the basic assumption
on which Robert seems to build his entire case _valid_ in the context of
the language definition?  if not, his case be dismissed.

I was trying to point Robert in the direction of realizing that he had no
foundation for what he was claiming with such authority.  I still see no
foundation.  I see no realizing he has no foundation, either, and that is
what really irks me in this case.  to have the gall to conjure up some
elaborate "DAG" of arguments when his basic assumption is in question!

Robert, the HyperSpec is there for your perusal.  I suggest you use to
support your case.  I regret the need to play standards lawyer with you,
but you compel me to it.  (compare that with what you feel forced to.)

the question is: does ANSI Common Lisp (or CLtL2 for that matter) define
the semantics of `setq' at top-level, including whether the symbol should
be declared or proclaimed special?  this is a very specific question, and I
will hear no more about Robert's delusions of "vagueness" in response.

#\Erik