[r6rs-discuss] Last minute questions
benjisimon at gmail.com
Mon Jul 30 07:15:36 EDT 2007
Hey Everyone -
I feel like a student cramming for an exam :-). I have a few questions I'd
love to have answers to before I cast my R6RS vote. I've tried to find
these answers in the rationale document and by searching the mail archives,
but I can't seem to find what I'm looking for.
My intention here is not to start a flame war - I'm just looking for
reasoning behind decisions.
1. Why a standard library and not just dependence on SRFI's? Seems to me
like the SRFI process works well (I use them all the time with SISC) and
would allow us to shrink the spec significantly. Though I'm sure there's
there a good reason why the contents of the standard library are what they
are. What's that reason? A pointer to an existing discussion in the mail
archive would be perfect.
2. The module system used by
SISC<http://sisc-scheme.org/manual/html/ch10.html>(the one developed
for Chez Scheme) seems to provide less functionality than
the one provided for the in current R6RS draft. However, it seems more
composable (not sure if that's the right term) than the draft proposal. For
example, anonymous modules can be created in any lexical scope. The result
is that you can add functionality to the module system by creating macros.
This approach of creating a lightweight and flexible feature, with the
ability to enhance it by the user, seems more in keeping with the spirit of
Scheme. Is there a reason a more heavyweight / less flexible module system
was chosen? Again, a pointer to a conversation would be great. Also, please
correct me if my understanding of the module system is not correct.
Thanks for helping to make me an educated consumer.
My blog: http://benjisimon.blogspot.com
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the r6rs-discuss