--- Begin Message ---
On Thu, 2009-02-19 at 15:49 -0500, William D Clinger wrote:
> In my opinion, changes that can't gain the support of
> 75% probably shouldn't have been made in the first
> place. I believe the good parts of the R6RS would
> have little trouble gaining the support of 75%, but
> that may just mean I have more respect for the R6RS
> than some of those who proclaimed its perfection or
> voted to ratify it.
I dunno. I've been an outspoken critic of R6RS, and
I suppose I still am, but I find that the more I use
it the less I dislike it.
The module system is good and addresses a huge problem
in a useful and constructive way. My "objection" is
that having defined it, R6 didn't take full advantage
of it by moving functions into it and separating them
into multiple libraries. Throwing it out, especially
now, would be negative progress. Finishing or advancing
the good work begun would be more appropriate.
R6 Unicode support is more precisely conforming to the
Unicode standard and more compatible with how modern
Operating Systems treat text than I've seen in any other
language. These are good things. My arguments against
it aren't about its quality, they're about the requirements
being clumsily made in a way that results in all other
character repertoires and all possible extensions being
banned. There are adjustments to make, certainly, but
it would be pretty useless to undo good work that has
been done.
There's a compelling case for a scheme-oid lisp with a
"strings as immutable values" interpretation where
characters are just strings that happen to be short.
It would be cleaner semantics than R6, but such
a large semantic change and breaks so much existing
code that I can't imagine that it's likely to be a
reasonable direction to go.
Bear
--- End Message ---
_______________________________________________
r6rs-discuss mailing list
[email protected]
http://lists.r6rs.org/cgi-bin/mailman/listinfo/r6rs-discuss
