This question -- SVD for PCA: Why would one standardize the data matrix? -- was closed as a duplicate at 16:59:24 (nine minutes after having been posted). I can swear I saw it closed and without answers (that is when I wrote my comment). But then at 17:00:02 an answer was posted.
Answers cannot be posted to closed questions. So this is clearly due to some glitch or a loophole that allowed posting an answer within very short time after the question closure. I remember discussing a similar case in chat some time ago and one of the mods said that indeed it can happen.
I am curious what exactly the glitch/loophole is, and whether it is a bug or by design.
(Is it that if somebody opened the question page before the Q is closed and so can access the answer form, then they will be able to post an answer? So the SE software does not check if the Q is closed when receiving an A? Or it does check, but intentionally allows some leeway? Or is this leeway an unintentional consequence of some database updates taking time to propagate? Etc.)