On 8 Sep 2009, at 2:51 pm, Lynn Winebarger wrote:

> This is an extremely cogent explanation of why getting Scheme to
> support metacircular compilers, and not just interpreters, is
> important.  Much better than anything I've been able to express.

Indeed! Well done, Shiro!

Put another way, I think that failing to standardise an API to (eg)
parse RSS feeds in R7RS does not, in any way, imply failing to
standardise such an API elsewhere, perhaps in an SRFI, or just a
widely-used open-source RSS parser.

We just need to push for an R7RS that will foster the development of a
widely-used portable RSS parser, and the HTTP client library to find
some malformed RSS for it to parse :-)

So I want to see R7RS giving us a standard way to require libraries,
and an SRFI that standardises an HTTP client library (perhaps
implemented in terms of another SRFI that standardises a high-level
sockets API, perhaps implemented in terms of another SRFI that
standardises a Scheme interface to something at the level of BSD
sockets)...

Where we place the boundary between SRFI and Scheme Report is a matter
of organisation (who would like to be responsible for what) rather
than anything that really affects the power of Scheme as a programming
language, IMHO. Whether an implementation clams "R7RS + HTTP SRFI" or
just "R7RS" is irrelevant, if it gives us our portable HTTP client API.

ABS

--
Alaric Snell-Pym
Work: http://www.snell-systems.co.uk/
Play: http://www.snell-pym.org.uk/alaric/
Blog: http://www.snell-pym.org.uk/archives/author/alaric/




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

Reply via email to