14

Just looking for clarification of the off-topic close rules here.

My question How to remove indentation from one section? was just closed as off-topic. The comments indicate that at least one person voted for closure because it was answered in the comments.

I could see this as a valid close reason if the question was due to a typo or was otherwise extremely localized, but I think both the question and the answer are broad enough to be useful to other users. It's also about LaTeX, as required by the on-topic rules, and it's not subjective

I had invited the commenter to convert their comment to an answer as this would be the normal procedure on other stack exchange sites. I thought that since comments are not permanent it is more desirable to have the comment become an official answer, but instead my question was closed.

Of course, there may be other reasons that voters closed my question, and if it's off-topic for other reasons, I'm fine with it remaining closed, but I was surprised by the comment that it was closed for being answered.

3 Answers 3

7

The 'official' way is indeed to convert the comment into an answer, see How to end the "unanswered" status of really solved questions in the comments and Solved by comment -> how to flag. The person writing the answer can be either the original commenter, or the asker (along the lines of 'from the comment of @johnsmith I found the answer and it is as follows [...]), or a third person, that may or may not use the Community Wiki feature. All three options happen regularly.

However, on TeX.SE it is common that short answers are posted as comments, where the commenter feels that their solution is too simple for a full answer. See Why do people answer in comments? for some background on this practise - the main reasons are that it is untested, or that it is a near-duplicate (see question X but change Y into Z), or that it is a very small modification to the code of the OP (remove \par in the third line). In this case the commenter does not always want to write an actual answer, for the same reason that they originally chose to write the comment.

These questions are regularly closed as "solved in comments", as a way to mark them as being resolved. This is contrary to the Stack Exchange model of "comments are for clarifications and answers are for answers", but it works well for us, mainly because these comment solutions are usually correct and therefore answer voting is not necessary to determine quality, as is needed on other SE sites. This kind of comments can effectively be regarded as permanent - our moderators will not remove comments providing an answer, and commenters also generally do not remove their own comment if it has the answer.

It may happen that a question is closed too early and the commenter actually would have been interested in writing an answer. However, this does not happen very often I think, and in any case if the commenter wanted to write an answer then they should have done so in the first place. Such a closure does not mean that your question is off-topic. In that case (most) close voters will type "is off-topic" or "is not related to LaTeX" in the custom close reason box, instead of "solved in comments". This is unfortunately not shown in the closure box at the top of the question, but it is visible in the auto-generated closure comment below the question.

3
  • 2
    Would I see the custom close reason? On my question, the close reason just says "off-topic".
    – 2cents
    May 26, 2021 at 13:33
  • @Cecilia unfortunately the system does not show the custom reason in the box, but it does show what the user typed in the auto-generated comment below the question. I edited my answer to clarify this.
    – Marijn
    May 26, 2021 at 14:07
  • 4
    The effects of closing, as far as I can see, are: (1) it is "a way to mark them as being resolved" (quoting from above), (2) it prevents anyone from answering the question, (3) it can suggest to the OP or to other users that there is something wrong with the question and it is not welcome on this site. AFAICT the only people who benefit from (1) are those who care about statistics ("unanswered questions" etc), while future users are negatively affected by (2) and (3), so I've never been a fan of this practice of declaring "this question is off-topic because it has been answered already". May 30, 2021 at 7:50
12

I agree with your interpretation that a question about a problem resulting from a typo or failure to load a necessary package is fair game for closure, although "solved in comments" isn't very informative. It's easy enough to point out the nature of the problem/solution in a closing comment.

If the answer is found among the answers to another similar question, then pointing out the duplicate is preferable, But if there's no other evidence on the site, a brief answer is desirable.

Closing as "solved in comments" if the commenter has been asked to post an answer is at least bad manners.

2
  • 1
    Note that here the user that provided the solution in comments was the same user that cast the first close vote, so no "bad manners" in this case (unless you mean that the commenter should have honoured the request of the OP and by not doing so he displayed bad manners?). But in general you are of course correct that it is polite to wait if an answer might still be posted.
    – Marijn
    May 26, 2021 at 13:57
  • 2
    @Marijn -- In fact, I hadn't noticed the identity of that user. I think I've done the same, actually, but when I did, also posted a simple answer so that it would persist even if comments were removed. Not all people with questions ask them in the same way, or use the "correct" search terms, so I consider it useful to leave additional "bread crumbs" for later searchers, as long as they aren't exact duplicates. May 26, 2021 at 14:12
12

Closing as "solved in comments" is the worst thing to do to a question. If the question does not have enough votes, the closure might make it eligible to be roombad (https://tex.stackexchange.com/help/roomba) and all the knowledge will be lost. Further users with the same problem won't be able to benifit from it.

"Solved in comments" should only be used for problems others won't likely repeat like blatant typos.

4
  • Related discussion from last year: tex.meta.stackexchange.com/questions/8484/…. As far as I know the new Roomba requirements have not been implemented, so questions with more than 1 comment are still safe (for now).
    – Marijn
    May 26, 2021 at 13:46
  • 1
    @Marijn the 1 or more comments criteria only safes non-closed questions, that's why not closing will preserve them even if they are not answered, but the act of closing makes them susceptible for roomba May 26, 2021 at 13:51
  • I didn't realize that indeed questions that are closed like this are deleted if they have a score of 0 or less. That is a serious consequence (although many of the questions which are closed as answered in comments do have a few upvotes). Next time I vote to close like this I will upvote first.
    – Marijn
    May 26, 2021 at 14:04
  • 1
    @Marijn yes, voting helps a bit but is not ideal. A single downvote could still push the post over the treshhold and voting just to influence something else skews the usefulness this metric and might not be lasting in case of account deletion. May 26, 2021 at 16:36

You must log in to answer this question.

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