| In a message dated 10/6/2002 3:15:51 PM Central Daylight Time, a.rosta@lycos.co.uk writes: << > In the context of fuzzy logic, tall=1 either means he is the tallest >> I took these as two posssible bases for a multivalent logic (maybe even fuzzy). "Tall enough" looks like a function which would work on either of these (or any other) logic (apparently with a longer T, a shorter F and a steeper slope. << > Another issue with ja'a + CAI is that CAI can already float wherever it > likes in a bridi, and could modify the ".i" as well, so including ja'a in > there is totally redundant. What's a good convenient but unvague way to say "very X", then, in contrast to "pretty X-ish"? >> I still think it is a good idea to keep functions and comments separate -- for getting truth evaluations if nothing else. << > I hope you don't mean that ja'axipibi melbi is less than melbi! That is what I meant. > I should think it means the le melbi is .8 on [0, 1] of melbi. Obviously I disagree, but if there is some better way to indicate 'sort-of', then I don't care that much. Maybe the best way for me to proceed would be to outline a scheme using experimental cmavo, & then we can see if we can find ways to do it with existing resources with comparable ease. >> Personally, I would advise sorting iut the various things you are about before going into this. "sorta" is a favorite function of mine and the dispute between the two of you seems to be between truth values and event quantities (or, in xod's case apparently, quantities in the event) -- or maybe just different ways of setting up the truth values. |