From: Brian Harvey <[email protected]>
Subject: Re: [r6rs-discuss] What's under the hood?
Date: Thu, 24 Sep 2009 23:57:44 -0700

> This position is counterintuitive, I know, because I'm advocating putting in
> WG1 an abstraction that, in WG2, would be written in Scheme using lower-level
> tools that (my idea of) WG1 just won't have.  It might well turn out that you
> can't write a really efficient WG1 interpreter in WG1-Scheme, and that
> implementations of WG1-Scheme will routinely be written in WG2-Scheme.  That
> won't bother me, as long as you can write a metacircular WG1 interpreter,
> even if it's inefficient.

This is a reasonable idea.   I feel I can live with that
Scheme if it is accompanied by rich srfis.
Some srfis may require primitives that can't be expressed
in the core Scheme (e.g. network srfi may require socket
interface), but it's ok as far as the feature set is 
implementable in reasonable effort of each implementor.

What I care is that such a Scheme still needs to guarantee
certain level of portability and interoperability.
Case-(in)sensitivity is one of issues related to them.
So #!fold-case and #!no-fold-case are, although ugly,
needs to in or around the core, for example.
(Or we can debate again to choose one; it's not the
topic of this post.)

Character code is another portability/interoperability
issue.  I'm ok for an implemenatation to choose whatever
character set it's suitable, but we need to agree that
if we exchange code, how those character should be
interpreted.  

But then, it occurred to me that even those issues can be
delegated to "portability srfi"... a piece of code may
declare that it is portable if the implementation satisfies
this and this portability srfi, like "srfi-xxx character as
unicode codepoint" or "srfi-yyy character as extended
grapheme cluster".   Or "srfi-zzz metasyntax to switch
case sensitivity".  (I'm stretching myself here.)

Will that lead to dependency headache between srfis?  I'm
not sure.   If many srfis require one particular srfi, that
one will be de facto standard.   The danger is probably
having equally popular srfis that cannot coexist.
I guess string mutability and "what is a character?"
problem have such danger.

--shiro

_______________________________________________
r6rs-discuss mailing list
[email protected]
http://lists.r6rs.org/cgi-bin/mailman/listinfo/r6rs-discuss

Reply via email to