[This question](http://mathoverflow.net/questions/239783/normalization-of-complete-intersection) (which I did not ask) is whether the normalization of a local complete intersection domain is Gorenstein.  Why was this closed as "off-topic"?  Especially when [this one](http://mathoverflow.net/questions/68246/cohen-macaulay-domain-with-non-cohen-macaulay-normalization), which asks whether the normalization of a Cohen-Macaulay domain is CM, was not.  I could understand closing as "duplicate", because the counterexample to one works for the other (does that count as duplicate?), but "off-topic", really?

Maybe the specialists think it's too trivial to be a research question.  Of course it's not a research question *in itself*, but I can quite easily imagine how one might come up with such a question incidentally (while doing research on an adjacent subject), and not have anyone around to ask, so MO would be the natural place to go to.  (And it's reasonable first thought to think that normalization will tend, if not to improve, at least to preserve, singularities along the CM/Gorenstein/CI line.)  At any rate, it's certainly not homework.  Also, the user in question seems to ask reasonable questions in general (although their wording could often be improved and, perhaps as a result, they have not received much attention).