Some thoughts on Motion 19 (still under vote)
P-1788,
As you vote on Motion 19, you might want to consider that
the "implicit" data type can encompass more than mid-rad.
In particular, if I recall correctly, one of the items the
small (not posted to this mailing list) group discussed prior
to formulating the motion was minimal standards for arithmetics
that were not based on IEEE 754, such as multiple precision
arithmetic and variable precision staggered arithmetic. Such
arithmetics, implemented well, can give SHARPER results, over
an expression that contains more than one operation, than what
we might be envisioning for an explicit interval data type.
However, one could also argue that these arithmetics are still
under development, one could argue that all we need is an
exact or accurate dot product instead, etc. (However, I've seen
implementations and useful results from these arithmetics.) Also,
you might want to vote against Motion 19 if you either think
such arithmetics should be outside the realm of the present
standard or if you think motion 19 is not a good vehicle for
including them.
Baker
--
---------------------------------------------------------------
R. 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
---------------------------------------------------------------