>The real point I think is to
>point out how easy it is to make statements in natural
>language that are subtly ambiguous. Right. There is another current example at
https://www.cut-the-knot.org/htdocs/dcforum/DCForumID11/3.shtml
>Natural language is
>fuzzy compared to say predicate logic.
I am sure this is why we humans use it in the first place. There is an interesting and quite relevant book Words and Rules by S. Pinker. Following rules all the time would be too hard on the brain. But remembering (as opposed to following rules) depends on other memories which are not universally shared; hence frequent ambiguities.
>I just find these sorts of mind benders fascinating. On one
>hand we are trying to make computers “think,” but on the
>other hand we don’t want computers to make mistakes.
You allow here the kind of ambiguity which is the subject of the discussion. We would not of course envisage a computer that returns 5 to a 2×2 = ? querry. On the other hand, it is easy to program a computer to be in a frivolous mode wherein it knowably returns 5 as an answer.
>This
>kind of mind bender seems to point out that we can’t have it
>both ways…
Or at least very difficult. But there is say fuzzy logic which would allow some leeway in terming a mistake mistake.