Subject: Re: XEXPR needs Schemers' help
From: rpw3@rigden.engr.sgi.com (Rob Warnock)
Date: 31 Dec 2000 04:34:48 GMT
Newsgroups: comp.lang.scheme
Message-ID: <92md18$jmh3t$1@fido.engr.sgi.com>
Ray Blaak  <blaak@infomatch.com> wrote:
+---------------
| Why bother with any of these? Scheme parsers are cheap, and could easily be
| hooked into an XML contents stream. All these XML transformations are simply
| artificial work.  Why not just:
|   <scheme>
|     (define square (x) (* x x))
|   </scheme>
| and leave it at that?
+---------------

Indeed. Why not [with corrected syntax] just:

      (define (square x) (* x x))

Who needs XML?   ;-}

Look, the point is that the XML/XHTML (and SGML before them)
communities keep re-inventing fully-parenthesized notation, but
with fatter and fatter parentheses, because "parentheses are ugly".

Why? I dunno. Seems silly to me. [Which in turn insprires me to be
somewhat silly when cramming Scheme's nicely-rounded syntax into
XML's square -- or worse, starfish-shaped! -- hole.] But *if* you're
going to re-invent Scheme or Lisp in XML syntax, don't do a half-assed
job! Get it right. Model the *actual* syntactic structure of Scheme,
not just a wrapper that says "(with-input-from-string (eval me))".

All by itself, Scheme is a perfectly fine language for writing "active"
web pages. Among others, see:

	<URL:http://brl.sourceforge.net/brl_1.html>
	<URL:http://brl.sourceforge.net/brl_4.html>

	<URL:http://www.cs.auc.dk/~normark/scheme/index.html>
	<URL:http://www.cs.auc.dk/~normark/scheme/laml-motivation.html>


-Rob

-----
Rob Warnock, 31-2-510		rpw3@sgi.com
SGI Network Engineering		http://reality.sgi.com/rpw3/
1600 Amphitheatre Pkwy.		Phone: 650-933-1673
Mountain View, CA  94043	PP-ASEL-IA