arithmetic issues Aubrey Jaffer (21 Oct 2005 14:53 UTC)
Re: arithmetic issues John.Cowan (21 Oct 2005 15:59 UTC)
Re: arithmetic issues bear (21 Oct 2005 16:39 UTC)
Re: arithmetic issues John.Cowan (22 Oct 2005 02:03 UTC)
(missing)
(missing)
(missing)
(missing)
(missing)
Re: arithmetic issues Marcin 'Qrczak' Kowalczyk (23 Oct 2005 20:24 UTC)
Re: arithmetic issues Thomas Bushnell BSG (23 Oct 2005 20:30 UTC)
Re: arithmetic issues Marcin 'Qrczak' Kowalczyk (23 Oct 2005 22:25 UTC)
Re: arithmetic issues Thomas Bushnell BSG (23 Oct 2005 22:30 UTC)
(missing)
(missing)
Re: +nan.0 problems bear (24 Oct 2005 06:04 UTC)
Re: arithmetic issues Marcin 'Qrczak' Kowalczyk (21 Oct 2005 17:15 UTC)
Re: arithmetic issues Taylor Campbell (21 Oct 2005 20:24 UTC)
Re: arithmetic issues Thomas Bushnell BSG (21 Oct 2005 20:32 UTC)
Re: arithmetic issues Alan Watson (22 Oct 2005 00:26 UTC)
Re: arithmetic issues Marcin 'Qrczak' Kowalczyk (22 Oct 2005 00:45 UTC)
Re: arithmetic issues Aubrey Jaffer (22 Oct 2005 01:17 UTC)
Re: arithmetic issues Thomas Bushnell BSG (22 Oct 2005 01:22 UTC)
(missing)
(missing)
(missing)
(missing)
Re: arithmetic issues Bradley Lucier (23 Oct 2005 19:46 UTC)
Re: arithmetic issues Aubrey Jaffer (23 Oct 2005 20:10 UTC)
Re: arithmetic issues Aubrey Jaffer (23 Oct 2005 19:54 UTC)
Re: arithmetic issues Jens Axel Søgaard (23 Oct 2005 20:01 UTC)
Re: arithmetic issues Aubrey Jaffer (23 Oct 2005 20:50 UTC)
Re: arithmetic issues Thomas Bushnell BSG (23 Oct 2005 21:12 UTC)
Re: arithmetic issues Marcin 'Qrczak' Kowalczyk (23 Oct 2005 22:31 UTC)
Re: arithmetic issues Thomas Bushnell BSG (23 Oct 2005 22:33 UTC)
Re: arithmetic issues Marcin 'Qrczak' Kowalczyk (23 Oct 2005 22:50 UTC)
Re: arithmetic issues Thomas Bushnell BSG (23 Oct 2005 22:57 UTC)
Re: arithmetic issues Marcin 'Qrczak' Kowalczyk (24 Oct 2005 00:53 UTC)
Re: arithmetic issues Thomas Bushnell BSG (24 Oct 2005 01:05 UTC)
Re: arithmetic issues Marcin 'Qrczak' Kowalczyk (24 Oct 2005 01:45 UTC)
Re: arithmetic issues Taylor Campbell (24 Oct 2005 02:00 UTC)
Re: arithmetic issues Marcin 'Qrczak' Kowalczyk (24 Oct 2005 02:08 UTC)
Re: arithmetic issues Taylor Campbell (24 Oct 2005 02:14 UTC)
Re: arithmetic issues Marcin 'Qrczak' Kowalczyk (24 Oct 2005 02:27 UTC)
Re: arithmetic issues Taylor Campbell (24 Oct 2005 02:45 UTC)
Re: arithmetic issues Alan Watson (24 Oct 2005 02:13 UTC)
Re: arithmetic issues Taylor Campbell (24 Oct 2005 02:22 UTC)
Re: arithmetic issues Alan Watson (24 Oct 2005 03:19 UTC)
Re: arithmetic issues Thomas Bushnell BSG (24 Oct 2005 02:01 UTC)
Re: arithmetic issues Aubrey Jaffer (24 Oct 2005 02:27 UTC)
Re: arithmetic issues Alan Watson (24 Oct 2005 03:14 UTC)
Re: arithmetic issues John.Cowan (24 Oct 2005 05:37 UTC)
Re: arithmetic issues Per Bothner (24 Oct 2005 07:05 UTC)
Re: arithmetic issues Marcin 'Qrczak' Kowalczyk (24 Oct 2005 07:58 UTC)
Re: arithmetic issues Thomas Bushnell BSG (24 Oct 2005 08:05 UTC)
Re: arithmetic issues Alan Watson (24 Oct 2005 08:25 UTC)
reading NaNs Aubrey Jaffer (24 Oct 2005 15:35 UTC)
Re: reading NaNs Per Bothner (24 Oct 2005 17:35 UTC)
Re: reading NaNs bear (24 Oct 2005 19:23 UTC)
Re: reading NaNs Marcin 'Qrczak' Kowalczyk (24 Oct 2005 18:17 UTC)
Re: arithmetic issues bear (24 Oct 2005 06:13 UTC)
Re: arithmetic issues Taylor Campbell (24 Oct 2005 06:27 UTC)
Re: arithmetic issues Thomas Bushnell BSG (24 Oct 2005 07:49 UTC)
Re: arithmetic issues bear (24 Oct 2005 16:41 UTC)
Re: arithmetic issues Thomas Bushnell BSG (24 Oct 2005 07:49 UTC)

Re: arithmetic issues Bradley Lucier 23 Oct 2005 19:45 UTC

On Oct 23, 2005, at 2:39 PM, Thomas Bushnell BSG wrote:

> Aubrey Jaffer <xxxxxx@alum.mit.edu> writes:
>
>
>>  | From: Thomas Bushnell BSG <xxxxxx@becket.net>
>>  | Date: Sat, 22 Oct 2005 17:47:25 -0700
>>  |
>>  | Aubrey Jaffer <xxxxxx@alum.mit.edu> writes:
>>  | > ...
>>  | > That still prevents an implementation from displaying
>> information
>>  | > about what type of NaN was returned.  Such information could be
>>  | > helpful to find the call which generated the NaN.
>>  |
>>  | Huh?  How does it prevent such?  We *could* mandate a readable
>>  | written representation for NaNs without manding that printing a
>> NaN
>>  | should produce that representation, since it would still be
>> allowed
>>  | to signal an error.  (And then, once it is signalled, it could
>>  | print *anything it wants*.)
>>  |
>>  | Moreover, nothing prevents the mandated written representation
>> from
>>  | optionally including implementation defined contents, if that
>>  | should be useful.
>>
>> When different NaNs are returned depending on the circumstances
>> creating them, I would like my implementation to display them like
>> this:
>>
>>   #<not-a-number expt>
>>
>
> Sure, that seems fine.  We could mandate that as the readable written
> representation!

If "expt" means exponent, then you don't get any more information---
in IEEE arithmetic, the exponents of all NaNs of a given precision
are the same.

Brad (who doesn't really want to get into this, ...)