In CRAN Repository Policy they write:
In principle, packages must pass R CMD check without warnings or significant notes to be admitted to the main CRAN package area.
The term "significant notes" appears a bit vague. Any idea what do they mean by this in practice? Or perhaps the question should be turned, what is an "insignificant note" in R CMD check allowing publication on CRAN?
I have a feeling that "significant notes" might be those they mention indirectly (package size and processing time of examples). I still feel a bit confused by this term. Hadley Wickham explains that each note has to be manually checked by a person and therefore notes should either be eliminated or mentioned in the submission comments. The whole point is to save the time of volunteers, which is "CRAN’s most precious resource", they write. Following this instruction, it would be nice to know which R CMD notes will definitely lead to the rejection of a package.