I think you are replying to a different argument than the
one I was trying to make.

I wasn't suggesting that _. be 'above that' in the sense of
following certain J-defined rules independent of the
IEEE spec.  That would be OK but if it's too hard,
let it go.

I am happy with all the implementation decisions you have made.
I am particularly happy with raising error when _. would
be created.

My only objection is to using the name 'NaN'.  Maybe you use
a NaN for _., but I see no reason to pollute the documentation
with that term.  I AM suggesting that the language spec be
'above that', because NaN is a new and unnecessary concept
within J numbers.

And _. IS a number.  qbeta tells me so:

   9!:14''
j602/beta/2008-02-22/22:30
   3!:0 (_.)
8

Don't change a line of code.  Do change the doc of 128!:5 to say
it checks for the presence of _. rather than of NaN.  Do change
'NaN error' to either 'domain error' (preferred) or '_. error' .

The only appearance of NaN in the docs should be the one in _.,
which is

  The indeterminate _. is provided to aid in dealing with
  NaN (not a number) in data from external sources...

This makes clear that NaN is IEEE, not J.  Occam's razor.

Henry Rich

> -----Original Message-----
> From: [EMAIL PROTECTED] 
> [mailto:[EMAIL PROTECTED] On Behalf Of Eric Iverson
> Sent: Sunday, February 24, 2008 12:08 PM
> To: Beta forum
> Subject: Re: [Jbeta] Use of the name 'NaN' deprecated
> 
> Your saying _. is indeterminate and has nothing to do with 
> the IEEE spec of 
> Nan doesn't make it so. Our conclusions are that trying to 
> make J 'above 
> that' is a fools game that has a very high performance cost 
> for all fp 
> operations. The hard, unavoidable fact is that J uses the 
> IEEE fp spec as 
> provided by the hardware and compiler. To do otherwise is not 
> worth the 
> effort and has far too high a cost.
> 
> If you are interested and concerned about _. (whatever it is 
> called), I 
> strongly suggest you take a look at what is in the current 
> beta as that is 
> what is going to be in the release.
> 
> ----- Original Message ----- 
> From: "Henry Rich" <[EMAIL PROTECTED]>
> To: "'Beta forum'" <[email protected]>
> Sent: Sunday, February 24, 2008 11:47 AM
> Subject: [Jbeta] Use of the name 'NaN' deprecated
> 
> 
> > Roger and Eric have referred to NaN in their messages.  I suggest
> > that this usage should be avoided.
> >
> > NaN is meaningful only in reference to the IEEE floating-point
> > spec.
> >
> > J is above that.  J deals with numbers.  Floating-point is an
> > implementation detail that should not be alluded to in the
> > description of the language.
> >
> > _. is indeterminate.  It is not NaN.  For one thing, it is
> > a number (at least it used to be - I haven't had the courage
> > to move to rbeta yet), while NaN is explicitly not a number.
> > And, there are many values and kinds of NaN, but only one
> > indeterminate.
> >
> > So, NaNs in external sources produce unpredictable results.
> > Use of _. produces unpredictable results.  But they are
> > not the same things.
> >
> > Henry Rich
> >
> >> -----Original Message-----
> >> From: [EMAIL PROTECTED]
> >> [mailto:[EMAIL PROTECTED] On Behalf Of Eric Iverson
> >> Sent: Sunday, February 24, 2008 9:37 AM
> >> To: Beta forum
> >> Subject: Re: [Jbeta] issues not yet resolved
> >>
> >> I am sorry to say that we won't get fixes for these two
> >> problems into this
> >> release. We have just run out of time and need to get this
> >> release out so we
> >> can clear the decks for the next round of activities. The NaN
> >> problem turned
> >> into a surprisingly difficult mess and took much longer 
> than expected.
> >>
> >> Your two bugs will be at the top of the list for 602.
> >>
> >> ----- Original Message ----- 
> >> From: "bill lam" <[EMAIL PROTECTED]>
> >> To: "Beta forum" <[email protected]>
> >> Sent: Saturday, February 02, 2008 12:05 AM
> >> Subject: [Jbeta] issues not yet resolved
> >>
> >>
> >> > [Jbeta] for-loop, continue, switch-case  30 Sep 2007
> >> > 
> http://www.jsoftware.com/pipermail/beta/2007-September/002383.html
> >> >
> >> > [Jbeta] m&i. and boxed unicode 30 Sep 2007
> >> > 
> http://www.jsoftware.com/pipermail/beta/2007-September/002382.html
> >> >
> >> > Will they be fixed?
> >> >
> >> 
> ----------------------------------------------------------------------
> >> > For information about J forums see
> >> http://www.jsoftware.com/forums.htm
> >> 
> ----------------------------------------------------------------------
> >> For information about J forums see
> >> http://www.jsoftware.com/forums.htm
> >
> > 
> ----------------------------------------------------------------------
> > For information about J forums see 
> http://www.jsoftware.com/forums.htm 
> ----------------------------------------------------------------------
> For information about J forums see 
> http://www.jsoftware.com/forums.htm

----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm

Reply via email to