Re: Do we have implicit semi-infinites...?
On 2012-03-15 11:19:51 +0200, Svetoslav Markov wrote:
> As I said before, if an interval in mid-rad form is such
> that, say rad >= 1/2 |mid|, then according to common
> sense, such an interval is meaningless and useless.
> (Instead of 1/2 any other convenient fp constant between,
> say, 1/10 and 1/2 can be used.)
>
> This is because mid-rad numbers are used to model
> approximate numbers and the latter are useless when
> the error (rad) is too large or the mid-point is of unknown
> sign. Hence, infinite mid-rad intervals and mid-rad intervals
> containing zero are practically useless.
>
> Negative radii correspond to improper (Kaucher) intervals.
> If Kaucher intervals are used, then a meaningful
> condition becomes, say | rad | < 1/2 |mid |.
>
> IMO if |rad| >= 1/2 |mid|, then used should be
> notified that there is somethng wrong with his/her
> computations.
I don't think there should be any notification. IMHO, the arithmetic
should be closed, i.e. support Entire, and an implementation would be
free to return Entire if the radius is considered to be too large for
the implementation. But you can regard Entire as a notification.
--
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)