From time to time one runs across answers here that offer nothing beyond a bare answer to the question taken in a narrow sense. In particular, there's no general explanation, no attempt to show a larger pattern by giving good and bad examples, nor anything beyond the simplest possible answer to only the question asked. (Sometimes, only the title of the question actually receives the minimal answer, such as this wondrous sampling, which at least is honest enough to admit to its failings.)
On most of Stack Exchange, and especially on the big original, Stack Overflow, the policy is that attempts to answer the question should not be deleted, but that those don't do a good job should be downvoted — and, if lacking an explanation, usually downvoted heavily. And I do not think that's a bad idea in general (I've voted Looks OK on probably hundreds such answers on SO alone), but ELL has some unusual circumstances. Here, most of the time the questions and answers are more precise and narrower in their application. So while there's some value in receiving code that just works, even with a very limited understanding of its function, there's much less value in being told which of four choices one should have picked on an exam, and nothing more. Our stock in trade here is not working code (and the understanding of it to follow), but simply understanding.
To put it another way, by the time they can use ELL, learners have already mostly graduated beyond the simplistic rote memorization that these answers provide in such small quantities.
This phenomenon is especially bad on word and phrase requests ("I often use 'Holy mona lisa!'."), but it can show up even on questions about verb conjugation ("The second sentence is not Standard English, and not acceptable.") Hot Network Questions are especially prone to these ('I would use the past tense verb "been" Carrie has been at the airport for two hours.'), and sometimes get more than one ("If you want to keep most of the words from the original sentence, I would use: Carrie has been at the airport ever since she arrived two hours ago."). Sometimes reviewers delete these; sometimes ♦ moderators do; sometimes different reviewers decide not to delete them. Sometimes they get accepted before they can be deleted, or just get more upvotes for being correct than can be countered by the downvotes they receive for lacking useful explanations. So it seems to me we should work out consensus for how to handle these, rather than relying on our current informal and diverging approaches.
Submitting Answers that merely answer the question is a similar discussion, but appears to be more from the perspective of answerers and voters, rather than 20k delvoters, 2k Low Quality Posts reviewers, and flaggers. Nor does it really deal with the bottom-of-the-barrel answers discussed here, which not only lack authoritative references, but, crucially, just don't have anything generally applicable.
