alt Hacker NewsHey, what about this idea? redefine interval representation, such that [a, b] means the same if a<b; but if b<a, it means [-∞, b]U[a, +∞]. Then your example would become 1/[-1,2]=[0.5,-1]
Hey, what about this idea? redefine interval representation, such that [a, b] means the same if a<b; but if b<a, it means [-∞, b]U[a, +∞]. Then your example would become 1/[-1,2]=[0.5,-1]
That's a very cool idea :) It was proposed as far back as 1968 (!) in a paper by none other than the legend of floating point himself: Wiliam Kahan https://interval.louisiana.edu/historical-preprints/1968-Kah...