Re: Motion: Number format (Motion 33?)
On 2012-04-14 21:09:55 -0700, Dmitry Nadezhin wrote:
> > > If we require these properties from numeric types, should not we
> > > formulate similar requirements to interval datatype.
> >
> > I think so.
>
> What do you think about formulation of requirements to F and
> requirements to T in the same motion ?
I prefer simple motions than larger motions with requirements on
different subjects (even though they look similar). Large motions
increase the risk of disagreement.
> > > 2) Interval [0,0] mut be in T (as A3)
> >
> > I'm not sure. Are there operations that would return this interval
> > explicitly? If neg() is not in the standard, [0,0] must be in T so
> > that the user can write [0,0] - x to get something equivalent to -x.
>
> Can anybody think of interval type T that doesn't contain [0,0] ?
One may also wonder if it could be useful to design a new interval
type T that doesn't contain [0,0], e.g. for testing purpose (to
detect bugs in user-side algorithms or in the implementation).
I don't think so, but it isn't easy to answer. For instance, in
MPFR, I added a generic test using the fact that the exponent range
doesn't have any specific requirement (in particular, it doesn't
necessarily contain 0).
--
Vincent Lefèvre <vincent@xxxxxxxxxx> - Web: <http://www.vinc17.net/>
100% accessible validated (X)HTML - Blog: <http://www.vinc17.net/blog/>
Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon)