Re: Motion 31 draft text V04.4, extra notes
Nate Hayes wrote:
> the comparison of "overflown" intervals
> [7,+OVR] \interior [1,+OVR]
> is necessarily false.
I disagree -- if we had the concept of +OVR, this comparison
would be UNORDERED, neither TRUE nor FALSE.
Note that, with decorations, we do have the ability to distinguish
between OVR and INF -- by means of the "bounded" decoration (or
the "unbounded" decoration, depending on which polarity works best).
However, we must be willing to swallow the possibility of UNORDERED
comparisons when overflown intervals are involved. This only happens
when BOTH comparands are overflown.
(Btw, "overflown" may be cute, but "overflowed" would be the correct
English word. We can of course always define "overflown" as a new
technical term!)
Michel.
---Sent: 2012-04-04 16:25:56 UTC