Re: Supporting set- and modal-paradigms
John Pryce wrote:
> (3) MOD intervals represented by [-oo,-oo] and [+oo,+oo] (Hayes 7.5)
> seem to be a problem, because they do not exist as separate
> objects in SET.
> (6) Operations on unbounded intervals. In SET there is nothing special
> about them: they are defined exactly as operations on bounded
> intervals. In MOD, Hayes 7.4 does not explain how the operations
> are defined, so I can't tell whether there is a problem here or
> not.
Regarding treatment of infinities and unbounded arithmetic in my paper:
although it might be nice engineering because it does not lead to any
exceptional conditions, there are problems with it. For example, the
solutions it produces are not unique, and group properties of the
modal/Kaucher intervals are lost. I don't recommend it is the best solution.
Nate