It seems that your change looks for words that may be duplicated (reused) across a range. That is, words that are too close to each other in the writing? That wasn’t the intent of this check, but someting interesting to consider as a future addition.

awesome addition to emacs. I have played a little with writegood-mode… May I suggest a small change in the duplicates section? I’ve been toying with

"\\b\\([[:word:]]+\\)\\([[:space:]]\\|\\s\\)+.\\{0,55\\}\\b\\1\\b"

(wrapped it between code blocks, I don’t know if it will get rendered correctly…)

to try to catch duplicate words like “then”, “therefore” and so on, in different lines. I guess there is a more neat way that does not catch TeX and similar, any idea?

Cheers,

Ruben