Ihor, > Does it make sense from the point of view of R code? > AFAIU, the current ob-R implementation converts lists into R tables, > which is not accurate? Would it make sense to convert Elisp lists into R > lists directly?
my "barely half a cent" would be that backwards compatibility here trumps, hmm, "logic". i suspect a note in the wiki page for ob-R stating that lists are converted to single-column tables would be sufficient. (partly because i think people writing code need to refer to such pages to understand these sorts of mappings.) (one could make the case that "no one" passes lists to R code; that could push *me* towards your suggestion.) cheers, Greg
