43

This is a public request to all Meta Stack Exchange users. There's a certain behaviour pattern that's... not great, but common, and this is basically a request to do this differently. More specifically: it's a request to use the system as it should be used, and as it's used by many other sites on the network.

It's this: People will completely and substantially answer a question to total satisfaction. Except, they'll do it in comments. That leaves the question answered, but not marked as having any answers or the possibility of one being accepted, left to be bumped by the Community for having 0 answers. It also means we can't use an important half of the voting mechanism to rate your answer, and we can't edit it to improve it. Until someone comes along to repeat basically what you wrote as an answer, which it should've been to begin with.

I don't know exactly what leads people to do this, though Why do some people answer in comments? leads to some insights. Answering a question is not what comments are for, even on a meta site. They're not used this way anywhere else on the network either, and whilst Meta is special, it isn't special in this way. Imagine if 10% of questions on your SE site of choice were answered in comments and that was the end of it, until someone felt like repeating that answer in an actual answer: that would be bad, and it's no better if it happens here.

So please: If you're answering the question, answer it in an answer, not in a comment.

Thanks for reading.

12
  • 34
    I usually post a comment instead of an answer when I think the question is a duplicate and I'm too lazy to find it. Likewise, I posted this comment for the same reason. I'm pretty sure it's a duplicate, but I'm too lazy to find it.
    – Mysticial
    Commented Apr 29, 2014 at 7:20
  • 8
    So what? If it's an answer, answer the question with an answer. Commented Apr 29, 2014 at 7:20
  • 8
    Nah, if it's a known duplicate, go find the duplicate. And if I don't have time for it immediately, I'll leave it as a comment and try to come back to it.
    – Bart
    Commented Apr 29, 2014 at 7:21
  • 4
    You know that even if a question's a suspected duplicate, you can still answer it, right? Commented Apr 29, 2014 at 7:21
  • 9
    Sure, I just won't. And even if it's not a duplicate, I often have first thoughts that might well become an answer at some point, but where I don't feel it deserves to be an answer just yet. In which case I might leave a comment which to some might be an acceptable answer.
    – Bart
    Commented Apr 29, 2014 at 7:22
  • 6
    If you're sort-of answering the question, just post it as an answer. Often your pseudo-answer is a pretty decent answer, and you can return to it later to improve it: that's how it works virtually everywhere else (except on sites like SO where comments are beyond anyone's capacity to police effectively). If it's the final answer everyone takes and accepts, you're removing our ability to enact any of the site's basic features with your answer, including downvoting it or editing it to correct it - unless we redo what you should have done to begin with. It's a practice that poorly impacts others. Commented Apr 29, 2014 at 7:25
  • 7
    Well, nope. I'll decide when I find a thought of sufficient quality to post it as an answer. You however are entirely free to make it one based on my thoughts in the comments. I won't hold it against you if you do. If you don't, you're just going to have to wait.
    – Bart
    Commented Apr 29, 2014 at 7:27
  • 5
    Ok, I can see where this could be getting off on the wrong foot in part: I'm not requesting people make answers they wouldn't otherwise, so go ahead and don't answer questions or dupes if you'd rather not. But what I'm talking about is cases where someone completely and fully answers a question to total satisfaction, but as a comment, and an actual answer need be no more substantial than what they wrote. If you do not wish to do that and would rather use comments anyway, I can't stop you - I'll just hope this request prompts others doing this to behave differently. Commented Apr 29, 2014 at 7:38
  • 9
    Another reason: While some nice guy is drafting his/her answer, someone comes and answers in comment before an actual answer is posted. So people are generally tend to upvote comment rather than his/her answer(who were drafting) because the commentator answered first! And even when the commentator converts his/her comment to an actual answer, people are tend to upvote that converted answer rather than the first one irrespective of the quality of the answers. #JustSaying #ExperiencedTooManyTimes
    – Himanshu
    Commented Apr 29, 2014 at 7:56
  • 12
    Links, or it didn't happen. Seriously, give me 5 questions that are completely and satisfactorily answered in the comments, and show me how that caused a problem such as having community bump the post. Commented Apr 29, 2014 at 13:15
  • 4
    @Won't, I can't tell if you're joking, but that mentality is the problem OP is highlighting. You think you're being non-chalant and informal. Cool, ya know? Don't care about rep or formality or getting max points. In reality, it makes it very difficult for others to use the site as it is meant to be used, for all the reasons he has been laying out. Commented May 26, 2015 at 4:55
  • 1
    "This is a public request to all Meta Stack Exchange users." - Why is this question specifically targeting "Meta SE"? This would seem to be a network-wide issue, particularly prevalent on StackOverflow IMO.
    – MrWhite
    Commented Sep 14, 2021 at 0:40

3 Answers 3

26

I could not agree more with the sentiment of this question!

As to why do people do it? It can't be laziness, because it takes equal effort to type the text and click the button.

I think it is either fear of downvotes or myopia (short sightedness):

  • Commenters may feel that their answer is too simple, lacking in depth, or perhaps it is untested and they don't want to risk their post being downvoted.

  • Commenters believe that the question is so "weak / under-researched" that it doesn't merit a "real answer".

  • Commenters may have the false belief that there are only two people involved in the interaction -- the asker and them. The truth is StackExchange sites are amazing resources for a growing population of researchers that is destined live on for many, many years. The pairs of eyes that hit a page are NEVER only 2.


What are the possible consequences?

Misinformation
Votes on comments only go up. So it is possible that less scrupulous users will naively upvote a commented solution which may be sub-optimal or even wrong. This may, tragically, snowball as more naive users continue to upvote the comment while users "in-the-know" are powerless to reduce the vote tally.

Premature Abandonment
Commenting answers can lead to premature question abandonment. I have seen this more than once (I won't say "a hundred times" because there is always someone who will say "show me 100 hyperlinks"). It could be: a question is asked, an answer is commented that fixes the issue, then other volunteers post answers as answers, but the OP is long gone and the others who posted answers have no way of being rewarded for their thoughtful effort. Boo.

Focal Point Disturbance
By design, solutions are to be posted as answers. Most readers know to review (at least) the top answers while researching to find the best solution and understand how/why they are appropriate. When researchers notice solutions in the comments, they need to spend (waste) time reading the comments to glean the additional solutions (which typically lack explanation) and compare the collection. Sometimes the comment thread is so long that it is collapsed and must be expanded to view which is a further hassle.

Rewarding Bad Questions
A fair percentage of volunteers want to be helpful. When some volunteers see that a basic question is asked (so basic that it can be solved with just a few keyboard strokes), they just jot down the solution as a comment. They want to help the asker, but the solution is so basic it is hard to add an explanation. When this happens, it would be better to downvote the question (if it is under-researched) and find a duplicate to close with (if available), perhaps post a link to the manual/documentation, and advise the asker to put forth more effort before posting. Generally, with these scenarios, the "greater good" is to not provide direct solutions, but to encourage the asker to "raise their game". Providing quick, commented solutions to these types of questions will teach the asker to return to the site and use it like a crutch at the first whiff of hardship (rather than search, research, and RTM) -- this chews up volunteers' attention/time which is better spent on more deserving questions.

Bad Role Modeling
There are site-culture ramifications too. When new users see veteran users posting solutions as comments, they assume it is okay. This perpetuates a problem that site designers explicitly ask not to be done.

vvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvvv enter image description here

^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

What do I see too often on StackOverflow?

Regex Pattern Droppers I like to swim in the regex waters, and frequently I see people going to the trouble of generating a regex pattern demo link and dropping it as a comment. These are seldom the most refined patterns. They mean to help, but they are going to no trouble to produce a high quality pattern. This behavior ends up cluttering the page, and future researchers will have to click on these commented links to find out if they are any good or not. If they were posted as answers, there could be upvotes, downvotes, and comments on them which would make things pretty clear and fair.

One-function Droppers Sometimes I see a snappy echo explode(',',$string); comment. These types of comments steal the thunder from other volunteers that take the time to post a fully explained answer. A fraction of these occurrences are used by FGITW users who try to prove that "they thought of the solution first". They chuck the solution as a comment, then feverishly type away at an answer to form a 1-2 punch that may sway the asker to select their answer because they helped first.


Meaning Well
Look, the majority of these people have their heart in the right place, but not their posts. There could be many reasons that they don't have the capacity to post a fully considered and explained answer. I understand that writing error-free and well-formatted code with a thoughtful explanation on a mobile device is somewhat tedious. I've had limited time because of something going on in my life. I've been impatient and didn't "feel like" writing out a full explanation. These situations should not stop you from being a contributor!

If you really want to help...

GET TO THE REVIEW QUEUES!!!!

The review queues are integral in keeping StackExchange sites clean. The higher your rep points, the more reviewing you can do. Reviewing is so easy, you can do it while you eat your cereal, ride the train, etc. SE sites want you to review -- this is why they offer so many badges for it. This simple contribution will better serve the online community than answering a question with a comment.

1
  • Yep fear of down votes is my choice (and still no accepted answer to this question!)
    – Ross
    Commented May 28, 2017 at 1:18
6

Yeah nah.

I think you're missing a vital ingredient that has persisted in Meta at least up until now - informality. Meta just has this certain je ne sais quoi that other sites don't have. As far as a lot of experienced users are concerned Meta reputation is useless, and they have more than they could ever spend (if there was something you could spend it on). They'll still answer your question, but not as a formal answer. The answer is right, but they don't care about it being voted up. Especially when it's the umpteenth question like it that they've answered.

Maybe that attitude will change a little now that the Meta split has happened. Time will tell.

Like others have done already I would have left this as a comment but it's a bit too verbose for that.

1
  • 7
    I agree it's more informal. I also find it reasonable they don't care about the reputation involved in getting upvoted. But the mechanisms available as part of a formal answer are important, because, y'know, that's why we have them at all, otherwise this place is just a forum. By leaving answers as comments, we don't get to apply that important stuff. There is not that much difference between typing in the comment box, or in the answer box. Commented Apr 29, 2014 at 11:25
-5

slugster was right: Meta is informal.

This question demonstrates the issue perfectly: it is not the sort of carefully written, focused, non-opinion-based question you would expect to see on a main site, so we should not expect comments and answers to follow the same standards either.

You must log in to answer this question.

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