| From: Sam TH <samth_at_ccs.neu.edu>
| Date: Thu, 16 Nov 2006 15:46:32 -0500
|
| On Thu, 2006-11-16 at 11:58 -0500, David Van Horn wrote:
| > For example, a comprehensive list library is a good SRFI
| > candidate. The community benefits by having a consistent,
| > complete, and portable library. Should this be part of R6RS?
| > No-- R6RS should give us the means for writing such a library,
| > not the library itself.
|
| I think this is a belief that is widely shared among members of
| this list, which is unfortunate. Consider the languages developed
| since 1980 that are still in wide use (C++, Java, C#, Python, Tcl,
| Ruby, Perl, CL, Haskell, SML, OCaml, etc). All of them (with the
| possible exception of SML) provide extensive libraries. Now
| consider the major Scheme implementations (I won't try to list them
| for fear of offending people). I believe that all of them provide
| libraries on par with SRFI 1.
|
| Are these facts coincidences? Of course not. Modern languages are
| more than just a syntax and semantics, they are the standard
| library as well. Scheme deserves a comprehensive library, not a
| minimal one.
Scheme has a comprehensive library: SLIB
(
http://swiss.csail.mit.edu/~jaffer/SLIB).
SLIB provides uniform base procedures and libraries across all R4RS
and R5RS Scheme implementations, enabling complex programs to be
written in implementation-neutral Scheme.
It has a procedural record module `record'
(
http://swiss.csail.mit.edu/~jaffer/slib_7.html#SEC204) and a SRFI-9
syntactic layer `define-record-type'
(
http://swiss.csail.mit.edu/~jaffer/slib_3.html#SEC39).
SLIB's `common-list-functions'
(
http://swiss.csail.mit.edu/~jaffer/slib_7.html#SEC206) is a
list-library comparable to SRFI-1, but without circular lists.
| Perhaps, for reasons of compromise between the editors, an
| extensive list library (or other library) cannot be agreed on.
| However, a useful standard library (covering records, lists, and
| many other things) should be a goal of RnRS.
Received on Thu Nov 16 2006 - 18:15:46 UTC