Thread Links Date Links
Thread Prev Thread Next Thread Index Date Prev Date Next Date Index

Re: -0 and +0



Vincent,

On 04/25/2013 08:08 AM, Vincent Lefevre wrote:
On 2013-04-25 05:50:08 -0500, Ralph Baker Kearfott wrote:
At this point, I will need to review all of our motions; perhaps
someone can help remind me. I vaguely recall we decided we would
have only one 0, and not -0 and +0 in the standard. (Of course,
this would not mean implementers could not use this 754 feature
in implementing 1788.)

This is a bit meaningless: What does "have only one 0" mean? At
which level? ...


I think John summarized it well: We're considering only
CLOSED intervals, and how do we interpret -0 and +0
in that context?  In any case, I am beginning to
regret bringing up the issue, which may be a bit
of a red herring, in the first place.  I thought I was
merely commenting on an innocuous issue of notation.

AFAIK, the main point that was decided about this is Motion 3,
which says:

   The P1788 Interval arithmetic standard defines intervals as
   closed and connected sets of real numbers.
   That means that ±Infinity may be used to denote an unbounded
   interval but are never considered as members of an interval.


YES!

This is at Level 1 and Level 2. Representation (Level 3) is another
matter, where you can't ignore that a number format may have signed
zeros...


and we may possibly not specify that much at level 3.  Isn't
level 3 still up in the air, anyway?

Best regards,

Baker


--

---------------------------------------------------------------
Ralph Baker Kearfott,   rbk@xxxxxxxxxxxxx   (337) 482-5346 (fax)
(337) 482-5270 (work)                     (337) 993-1827 (home)
URL: http://interval.louisiana.edu/kearfott.html
Department of Mathematics, University of Louisiana at Lafayette
(Room 217 Maxim D. Doucet Hall, 1403 Johnston Street)
Box 4-1010, Lafayette, LA 70504-1010, USA
---------------------------------------------------------------