more Re: [std-interval] list of functions for intervals

first.i.last at comcast.net first.i.last at comcast.net
Tue Jun 6 02:36:30 PDT 2006


 -------------- Original message ----------------------
From: Guillaume Melquiond <guillaume.melquiond at ens-lyon.fr>
> Le lundi 05 juin 2006 à 14:43 -0500, R. Baker Kearfott a écrit :
> 

[...]

> Indeed the cset of 0*inf contains -inf; hence e is forced to include the whole extended real line. 

[...]

> Without rounding (e.g. with rational bounds),
> interval e would have contained only positive and finite values. With
> rounding and without cset, interval e would have been [0,+inf).

This is an issue with the present state-of-the-art in csets, but I believe it to be fixable.  It derives from the concern over the topology of the number line near zero, but I believe there at least one sensible and rigorous definition that can be used to eliminate the issue and restore the IEEE-754 sense of rounding.

-- Lee Winter


More information about the Std-interval mailing list