In a message dated 3/19/2002 12:50:31 PM Central Standard Time, jjllambias@hotmail.com writes:I thought {du} was an infinite-place relationship. At least I'm (Yuck! ptui!) But so it does turn out to be. <>{li xy du me'o da pi'i >da} seems to work somewhat better (it still fails, but I am begining to >suspect that the parser doesn't do MEX any better than we do). The parser does MEX according to the rules. The problem is with the rules, not with the parser. I think it's {[li] xy du li me'o da pi'i me'o da}.> This is still not screwed up enough, i.e., the parser rejects both versions. <>1){roda de zo'u li da su'i de du li no} The not-quite-equivalence of sumti and operands can be quite annoying. {li da} is not grammatical. You could write it as: roda de zo'u li no sumji da de or, if you insist with MEX: roda de zo'u li mo'e da su'i mo'e de du li no> But the parser accepts this and gives it the right parse. I wonder what is different here from the previous case <>1'){roxy. zy. zo'u li xy su'i zy. du li no} mean ? {ro xy zy} is a single number. You need {roboi xyboi zyboi} in the prenex. Other than that, I think it works. (Though I prefer the {sumji} version.)> Both rejected, apparently {ro xy} is not recognized (though it should be as a case of "all the X's" even if not as a quantifier). The predicative forms all work, of course. |