[r6rs-discuss] [Formal] Allow compilers to reject obvious violations

From: Dr. Michael Pfiz <pfiz>
Date: Mon Feb 26 13:09:49 2007

> > Can you mention the part of R5RS that mandates a REPL?
>
> I didn't say that R5RS mandated a REPL. I said that the REPL had
> been removed from the spec. Note that LOAD, for example, is gone.


To cite the important parts of R5RS (Chapter 5.1??Programs):

"... Programs are typically stored in files or entered interactively to a
running Scheme system, although other paradigms are possible; ... Expressions
occurring at the top level of a program are interpreted imperatively; ..."

Nevertheless, this interactive use, which can't be provided without any kind
of read-eval-print-loop, is not dropped in R592RS, but somehow hidden in
Chapter 7.1 Rationale.


> In any case, one of the great things about Scheme and Lisp in general
> has always been their support for interactive, experimental use. If
> we have to patch an entire program before any part of it can be run,
> we will have lost another part of this interactive, experimental nature.

Without the load-procedure, the *experimental* character of interpreter use is
gone, but I try to write some lines on "libraries, extensions and safety" the
next days. The proposed standard seems focussed on compilers, but this is
just a personal impression from a casual scheme user.
Received on Mon Feb 26 2007 - 13:09:33 UTC