[scheme-reports-wg2] Reformulated numeric-tower ballot John Cowan (29 Apr 2014 03:49 UTC)
Re: [Scheme-reports] Reformulated numeric-tower ballot Jay Freeman (29 Apr 2014 06:51 UTC)
[scheme-reports-wg2] Re: [Scheme-reports] Reformulated numeric-tower ballot Taylan Ulrich Bayirli/Kammer (29 Apr 2014 07:55 UTC)
Re: [Scheme-reports] Reformulated numeric-tower ballot Jay Freeman (29 Apr 2014 08:37 UTC)
Re: [Scheme-reports] Reformulated numeric-tower ballot Peter Bex (29 Apr 2014 11:45 UTC)
Re: [Scheme-reports] Reformulated numeric-tower ballot Peter Bex (29 Apr 2014 13:18 UTC)
Re: [Scheme-reports] Reformulated numeric-tower ballot Peter Bex (29 Apr 2014 14:01 UTC)
[scheme-reports-wg2] Re: [Scheme-reports] Reformulated numeric-tower ballot Taylan Ulrich Bayirli/Kammer (29 Apr 2014 14:11 UTC)
Re: [Scheme-reports] Reformulated numeric-tower ballot Peter Bex (29 Apr 2014 14:25 UTC)
Re: [Scheme-reports] Reformulated numeric-tower ballot Alaric Snell-Pym (29 Apr 2014 22:13 UTC)
Re: [Scheme-reports] Reformulated numeric-tower ballot John Cowan (29 Apr 2014 22:54 UTC)
Re: [Scheme-reports] Reformulated numeric-tower ballot Alaric Snell-Pym (30 Apr 2014 12:59 UTC)
Re: [Scheme-reports] Reformulated numeric-tower ballot John Cowan (30 Apr 2014 16:38 UTC)
Re: [Scheme-reports] Reformulated numeric-tower ballot Peter Bex (01 May 2014 08:57 UTC)
Re: [Scheme-reports] Reformulated numeric-tower ballot Jussi Piitulainen (29 Apr 2014 13:58 UTC)
[scheme-reports-wg2] Re: [Scheme-reports] Reformulated numeric-tower ballot Taylan Ulrich Bayirli/Kammer (29 Apr 2014 16:14 UTC)
Re: [Scheme-reports] Reformulated numeric-tower ballot Devon Schudy (29 Apr 2014 15:38 UTC)
Re: [Scheme-reports] Reformulated numeric-tower ballot John Cowan (29 Apr 2014 17:04 UTC)
Re: [Scheme-reports] Reformulated numeric-tower ballot Devon Schudy (01 May 2014 13:02 UTC)
Re: [Scheme-reports] Reformulated numeric-tower ballot Alaric Snell-Pym (01 May 2014 13:11 UTC)
Re: [Scheme-reports] Reformulated numeric-tower ballot Takashi Kato (29 Apr 2014 20:07 UTC)
Re: [Scheme-reports] Reformulated numeric-tower ballot Perry E. Metzger (29 Apr 2014 22:48 UTC)
Re: [scheme-reports-wg2] Reformulated numeric-tower ballot Taylan Ulrich Bayirli/Kammer (30 Apr 2014 07:44 UTC)

Re: [Scheme-reports] Reformulated numeric-tower ballot Peter Bex 01 May 2014 08:47 UTC

On Wed, Apr 30, 2014 at 12:04:04PM -0400, John Cowan wrote:
> Peter Bex scripsit:
>
> > If an implementation doesn't support hash tables, it might not need
> > comparators either.  This is the same objection as I have with requiring
> > the full numeric tower.
>
> Where do you draw the line?  Lots of useful Scheme programs don't use
> vectors for anything: do you advocate making them merely optional
> as well?  If not, why not?

Good question.  I guess it's a little arbitrary.  I would say the main
difference is in the size of the standard.  r5rs or r7rs-small is a
reasonable size to implement in full, even for smallish implementations.
Requiring the whole of r7rs-large to be implemented seems overkill.

Since it's going to be so very large, I'm unsure whether *any* Scheme
(except, perhaps, for Racket) will implement every part of it.  That
means it's a good idea to avoid baking in too many assumptions about
which subset of r7rs-large is going to be implemented in any compatible
Scheme system.

> > I don't see why this has to be.  It will just exclude small
> > specialised implementations which would still like to support
> > a standardised library if it fits its intended use cases.
>
> There's no reason *not* to exclude small specialized implementations
> from a large standard.  That doesn't mean the implementations
> can't support libraries from -large if they want to:  I assume
> lots of libraries will work with Chibi even if they are not
> packaged with it.

If it still works with it, that's no problem.  But if things like "needs
built-in bignum support" are added to the spec, that may cause trouble.
And if it doesn't cause trouble, those parts of the spec that work with
Chibi turn out to not really need bignum support.  So why would you
arbitrarily require it for the entire spec, even if many libraries
aren't dependent on it?

> > For example, Chibi Scheme might decide to ship a few WG2 modules,
> > but you can compile it without bignum support.  Does that mean
> > it isn't WG2-compatible?
>
> If that ballot question passes, then yes, Chibi will not be R7RS-large
> compliant when compiled without bignum support.  Nothing wrong with that.

But it would only not be compliant because it was declared not to be,
for no real reason.

Cheers,
Peter
--
http://www.more-magic.net

_______________________________________________
Scheme-reports mailing list
Scheme-reports@scheme-reports.org
http://lists.scheme-reports.org/cgi-bin/mailman/listinfo/scheme-reports