Note: I do in no way want to say that the current moderators somehow do a "bad job" at moderating Physics SE. I think that they do a great work at, well, moderating this network and my full respect goes to them because I cannot even imagine how much work goes into this.

I often read comments something like this:

I have removed some comments that were trying to answer the question

posted by a moderator (of course, since they are as of my knowledge the only ones who can delete comments).

While I agree that comments are not for answering questions (they can be used, for example, to ask clarifying questions or give hints - I believe this does not count as an answer attempt), I disagree with how these comments are handled.

At least on my questions (when I see answer-like comments on other questions, they are most often already deleted (so I don't actually see them, but you get the point)), the comments often add valuable information. Sometimes, it is just some short explanation or links for further research, but they can still help a lot.

When the comments are deleted, this information is permanently gone. So I propose one of the following:

  1. The writers of answer-like questions should be notified that their comment should be an answer. They will have some time to either a) write an answer or b) if appropiate, add the content of their comment to an existing answer. If they don't do this in the given time, their comment will be deleted.
  2. Answer-like comments should not be deleted, but either hidden or moved. For example, discussions are often moved to chat; I think we've all seen this comment:

Comments are not for extended discussion. This conversation has been moved to chat.

I understand that this site and thus also comments have to be kept organized; but in the end, SE is a Q&A site. The goal is that users can ask questions and get answers - i.e. users help other users. I think that comments which do help should not be deleted (immediately or permanently).

  • 1
    $\begingroup$ Why the downvote? This is another thing that bothers me: Simply downvoting questions or answers without explaining so that the author can improve their question. The downvoter could also suggest an edit. $\endgroup$ – Jonas Jul 25 '20 at 20:55
  • 9
    $\begingroup$ Votes on meta are a bit weird, but a downvote here often means something like "I don't agree this is a problem" - i.e. they are not a judgement of the "quality" of your question but simply disagreement with the views expressed in it. $\endgroup$ – ACuriousMind Jul 25 '20 at 20:57
  • 8
    $\begingroup$ And come to think of it, I believe that's especially true when your question proposes a policy change ("Should ...?"). In those cases it's quite common for people to upvote or downvote the question based on whether they agree or disagree with the proposed change, so you can have a well-written proposal that accumulates downvotes simply because people are opposed to implementing it. $\endgroup$ – David Z Jul 25 '20 at 23:12
  • 3
    $\begingroup$ I think they should not be deleted, instead they are free resource for new answers based upon them. If someone actually answers a question in a comment, and another user writes an answer, using this comment (and also citing as a reference), then it is imho okay. $\endgroup$ – peterh - Reinstate Monica Jul 29 '20 at 22:25
  1. One problem with answer-comments you seem to not have considered is that they can be wrong. Answers can be downvoted, but comments can only be upvoted, so there is no proper mechanism to show that an answer-comment is considered wrong and the one-sided upvote mechanism may make it look more authoritative than it is.

  2. We do leave comments like "Please turn this comment into an answer" and users often do so - but after that's happened we'll delete both comments so you don't see any trace of this after the fact. The cases where you see the "answers in comments have been removed" treatment are likely high-traffic questions that already have answers where the comments did not add much value. You'll also see this more on high-traffic questions as a deterrent for the lots of visitors to not leave such comments, but write answers instead. I can't really speak across the actions of all mods but I think it's likely that your impression of what proportion of answer-comments are outright deleted vs. moved to chat or turned into answers is exaggerated because the deletion is intentionally more visible for this reason.

  • 1
    $\begingroup$ To 2. In this case, this then probably just was my impression. I apologize if I sounded rude or something like that. For 1., this problem could be fixed by converting the comment into an answer which can be downvoted - but apparently, as you said in 2, this also happens. $\endgroup$ – Jonas Jul 25 '20 at 21:00
  • 10
    $\begingroup$ @Jonas I thought you were perfectly polite - if all feedback on moderator actions was like this, it would be an improvement ;) $\endgroup$ – ACuriousMind Jul 25 '20 at 21:03
  • 6
    $\begingroup$ FWIW this answer is a good reflection of my comment-deleting practices as well. $\endgroup$ – David Z Jul 25 '20 at 21:27

While I agree that comments are not for answering questions (they can be used, for example...giving hints).

No, comments should not be used to give hints. If you have an answer, then just post an answer.

If you are giving hints for something like a homework question, then it is likely that the question isn't on topic anyway. Instead of leaving a hint, just flag/vote to close instead.

Comments are intended for either asking for clarifications or offering suggestions to improve the post. Anything else should either be put into an answer or not put anywhere on the post.

As for your main point here, I don't think we should leave invalid comments around, because then that makes them valid. The choice should either be to change our policy on what is allowed in a comment, or to actually enforce the current policy. I don't think we should be in some weird in-between where there is a subjective "yeah I know this isn't what comments are for, but maybe this time I can ignore that."

This would leads to inconsistent rulings. "I thought my comment was useful, why was it deleted? That comment over there wasn't deleted, and it definitely wasn't as useful as mine was." If the policy is that comments should be asking for clarification or suggesting post improvements, then that is all that should be allowed.

At least on my questions the comments often add valuable information....When the comments are deleted, this information is permanently gone.

Yes, this is exactly why users are told not to post answers in the comments. Just post an answer, that makes it way more useful.

For an example of a recently good answer that others would probably just comment, see v1 of this answer. Yes it is short, but it is an answer. So it should have been posted as an answer.

  • 3
    $\begingroup$ I think the reason why some people attempt to answer the question in a comment is because they feel that it could help the OP, but think that their comment is not detailed enough to be written as an answer. $\endgroup$ – Jonas Jul 26 '20 at 10:00
  • 3
    $\begingroup$ @Jonas Yes, that is probably why, but I don't think that reason makes it valid to make invalid comments. There is no requirement as to how "detailed" an answer needs to be. If you think "this isn't enough for an answer", then either give more detail so you can make an answer, or just leave it. $\endgroup$ – BioPhysicist Jul 26 '20 at 11:30
  • 4
    $\begingroup$ @Jonas The other piece to this is that our goal isn't to help OP -- rather, our goal is to assemble high-quality Q&A to help all people with similar questions in the future. Voting is an essential component of the high-quality part (which is what ACuriousMind touched on) and since comments are transient, making the hints answers is part of the in the future part that BioPhysicist is talking about when they say "Just post an answer, that makes it way more useful" $\endgroup$ – tpg2114 Jul 26 '20 at 17:05
  • $\begingroup$ Hi @BioPhysicist. Concerning my 1-line answer that you linked to, it indeed adequately answered v1 of the question. However, I later had to expand the answer considerably because the question was changed. $\endgroup$ – Qmechanic Jul 29 '20 at 13:59
  • $\begingroup$ @Qmechanic Ah ok, thanks for updating that for me! :) $\endgroup$ – BioPhysicist Jul 29 '20 at 14:00

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .