quasiquote vs. cons & friends

[Matthias Blume <blume@cs.Princeton.EDU>]

Here is a question for you, which bothered me for a long time:

Since quasiquote appears in the the ``derived expression types''
section of the report it seems like it has to be built on top of cons,
list, append, vector, etc.  Implementations of quasiquote as a macro
actually do this.  However, since the transcription of quasiquote into
simpler constructs is by no means obvious, and moreover since it can
vary from implementation to implementation I think quasiquote deserves
the status of a ``primitive expression type''.  Redefining any of the
global symbols (except quasiquote itself) shouldn't change
quasiquote's behavior.  (In VSCM quasiquote doesn't depend on the
values of any global symbol.)

What it the agreed-on opinion about this?  (I might ask: ``Is there
such a thing as an `agreed-on opinion' at all?'')

-Matthias