guile-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: JACAL, scm


From: Tom Lord
Subject: Re: JACAL, scm
Date: Thu, 4 Oct 2001 03:56:51 -0700 (PDT)


       Tom Lord <address@hidden> writes:

       > I think it is a good idea for GNU Emacs and the GNU extension language
       > to work well together.  They should be able to exchange data comprised
       > of basic types using `read' and `write' (on the Scheme side).  There
       > is no reliable way for the two programs to exchange simple data unless
       > () and #f are the same.

       Of course, but there is Jim Blandy's proposal for how to do it,

Which is not a very good proposal.  It causes more problems than it
solves.

Blandy's proposal takes an idea originally put forward for "how can we
execute very finicky R5RS code" and makes it the default top-level for
Guile (originally an R4RS Scheme).


       and it was adopted years ago as the way to do it in Guile, 

"Adopted" is a funny word here.


       and there is a nonzero set of people counting on it being done
       the original way planned.  Could you describe what's wrong with
       that proposal, in your view?

Yes, but not at all concisely.  Unless to echo the orginal "compatible
with lisp" requirement.  I can go on at indefinate length about why
"compatible with lisp" is a wise idea, and R5RS relatively
unimportant.  Doesn't seem like a good use of this list, though, given
the orientation and activity of the current maintainers.

You can get the gist of my view spread out in several recent posts to
comp.lang.scheme, though there are additional Guile-specific reasons,
the gist of which are spread out in previous messages to this list.

-t



reply via email to

[Prev in Thread] Current Thread [Next in Thread]