i like the idea of a program that can analyze a corpus and then identify when you have two many of the same "rare" words in a sentence, or too many "ultrarare" words in a document/paragraph.
It would be sort of a 'duplicate wild card' search.
Something like the old DOS 'star-dot-star' [*.*] that would define 'any' duplicates between period dots, for a whole document, without having to name particular words.
Also to be able to exclude simple common words like 'a', 'and', 'the', and so on.
It really seems quite simple, and Google already does it;
-'x' = 'any dictionary word' (the original DOS 'wild card' or star-dot-star *.*),
-find 'x', where 'x' repeats 2x and < (is less than [i.e. 'within (the borders of))] '.' --'.' (period dot/exclamation point/question mark) to 'same',
-exclude list; a, an, and, the...
A few variables might need to be written, to include every way a sentence can begin or end, as with quotes, and so on.
Well anyway, it's always fun to 'dream' or have a 'wish list'.
Forget about that new fad, the 'bucket list'; all I want when I get that far, is a dreamy-looking 'wish list' to fasten my eyes on.