Re: arithmetic issues William D Clinger (18 Jan 2006 05:54 UTC)
Re: arithmetic issues Alan Watson (18 Jan 2006 17:08 UTC)
Re: arithmetic issues Michael Sperber (18 Jan 2006 20:43 UTC)
Re: arithmetic issues Alan Watson (18 Jan 2006 22:15 UTC)
Re: arithmetic issues bear (18 Jan 2006 18:21 UTC)
Re: straw-man [was Re: arithmetic issues] Aubrey Jaffer (20 Jan 2006 16:13 UTC)
Re: straw-man [was Re: arithmetic issues] Alan Watson (20 Jan 2006 17:43 UTC)
Re: arithmetic issues Alan Watson (20 Jan 2006 17:48 UTC)

Re: arithmetic issues Michael Sperber 18 Jan 2006 20:38 UTC

Alan Watson <xxxxxx@astrosmo.unam.mx> writes:

> Would in not be simpler to allow implementations to do more or less what
> they want in the base language but mandate that the full tower be
> available in the library? For example, an implementation might implement
> fixnums and bignums in the base language (since for many problems
> ratnums, flonums, and complex numbers are not necessary) and then
> provide all of the rest of the tower in the library.

I don't know what exactly characterizes your notions of "base
language" and "library."  If "base language" means "what's needed to
build the other stuff," then that would clearly mean that the flonums
need to be in it.  If you mean "library" means something that an
implementation may or may not support, then see the rationale of the
SRFI why that wouldn't be a good idea.  If you think that SRFI 77
should really be made available in R6RS as a set of library modules
rather than just one big blob for organizational reasons, then I fully
agree with you---we just didn't have the methodology to do it when we
wrote the draft.  I'm hoping we might have it once we get to putting
stuff into R6RS.

--
Cheers =8-} Mike
Friede, Völkerverständigung und überhaupt blabla