16

I've seen comments on Super User Meta where users call it "toxic" or "bad behavior" to post an answer, and then delete it when it receives negative feedback or downvotes.

I've checked the Help Center and Meta but haven't found any official statement that explicitly forbids this. The only related guidance I see mentions vandalism or excessive deletion of useful posts.

Does Stack Exchange have an official policy about deleting one's own negatively received answers?
If not, is this mainly a community specific norm rather than a rule?

Clarification: I'm asking about this behavior on main Q&A sites, not on their Meta sites.


Edit: This example seems to have some background context I wasn't aware of. It's meant only as an illustration of the general behavior discussed, not a judgment of that specific situation.

Example (from the comment section):

The Rise and Fall of Super User Part 3 – Closing posts

A comment with two upvotes:

Let's be really serious. Submitting contributions and then deleting them because the user doesn't like the feedback they received is toxic behavior. I see this behavior from all sorts of users, from new users who don't know any better to users with hundreds of answers.

2
  • 4
    There's a lot of MSU, so... maybe a few examples for context would be nice for interpreting community norms Commented Oct 16 at 7:01
  • 3
    Man, there's a lot of... context behind the question and comments in question. Some of which I can't really touch on Commented Oct 16 at 7:50

6 Answers 6

24

You can get a badge by deleting your own answer with score -3 or lower. While some badges promote behaviour which is questionable in some circumstances, I don't think this is the case here.

If a user posts and deletes multiple answers with negative score, they'll eventually run into an answer ban, so any risk of 'abuse' is minimal. And in exceptional cases, moderators can have a private word.

16

I self-delete answers on occasion when there seems to be little point in keeping it:

  • if there's another, better answer than mine and I feel as though my answer detracts away from that
  • if I drastically misunderstand the OP or the OP changes their question so that it no longer applies to my answer to such an extent that it's not really helping
  • sometimes my answers might get comment-stormed from people who don't share my opinion, so I just shrug and delete the answer because it's obviously not productive

There's no site-wide policy on self-deleting and how to deal with it, on the whole the site isn't beset with this issue.

Any localized toxicity should be raised with site moderators or posted on the relevant meta for discussion.

9
  • 3
    "or the OP changes their question so that it no longer applies to my answer" At least in some (most?) communities, the OP mustn't change their question if it invalidates existing answers. Commented Oct 16 at 9:28
  • 2
    @Velvet I've seen this happen when an answer or comments turn out to unravel deficiencies in the question (or motives) and the OP then attempts to backtrack or change the narrative to something that puts them in a better light. Commented Oct 16 at 9:32
  • 6
    @Velvet mustn't isn't enforced and doesn't reflect reality. I've seen this a lot, and it's been discussed here a lot, too. In fact a long time ago I raised this feature request because of how often I saw people do this (intentionally or not). Just because you don't expect it to happen doesn't mean it isn't happening. :-) Commented Oct 16 at 13:17
  • 1
    @AaronBertrand I wasn’t aware of that aspect. On EE, "mustn’t" tends to be enforced quite strictly; especially once someone with a high rep has answered. Commented Oct 16 at 13:31
  • 1
    I actually would prefer people keep... valid but imperfect answers if better answers turn up. Sometimes a 'wrong' answer can have value in terms of... why its not correct as well. Commented Oct 17 at 2:22
  • @JourneymanGeek Sadly the only way to show something is incorrect is to comment (which are not permanent, are often ignored, and can lead to retaliation) or to down-vote (which can lead the author to pull the answer or, in cases where age/inertia has already led to up-votes, are just noise observed only by the morbidly curious who can and do expand vote counts). Commented Oct 22 at 16:20
  • @JourneymanGeek Having a mechanism to explicitly mark an answer as wrong is difficult, because "wrong" is as subjective as "right" (the asker "accepting"). But I'd love to eventually see something like the close banner, but on an answer, like "Potentially obsolete solution, see other answers on this page <answer link(s)> <voter(s)>. (Yes, I see the relation to the abandoned Outdated Answers project). Commented Oct 22 at 16:28
  • Well, I'd say comments work here - and cover most of what we need. Its also a valid use of comments - a request for clarification/provided clarification can say why an answer is wrong. Maybe as a 100+k user on the two sites I'm most active at, maybe I'm a little less worried about retaliation, and would leave such comments in place preferencially. Also a way to mark an answer as obsolete feels like community notes as used on social networking sites.... Commented Oct 23 at 0:25
  • @JourneymanGeek Yeah, I would like to see a way that everyone can feel comfortable marking an answer in some way that does not implicitly link it to a down-vote. Retaliation isn't a concern for you or me, but it is real and can affect newer users more harshly. Commented Oct 24 at 17:01
8

There's no real formal rule against it, but good reasons to discourage that. A good user's mindful of their posts - posts not for the sake of posting or the off chance of getting upvotes or other reasons, this fills the site with... less useful content.

So there's two possibilities here. A lower reputation user gets post banned, and.. is confused what they did wrong cause no one tried to tell them. Or a high reputation user who can't be post banned goes for volume, cause there's no downside to deleting a downvoted question, or one that's obviously incorrect.

Both of these are bad for us. We either lose a new user we could have educated (and maybe we need better ways to do this), or we have a high rep user who's not acting in the best interests of anything but accumulating reputation, and skipping out on the bad sides of poor quality posting. We don't catch everything so it might also encourage others to do it if there's no critique or votes. We also might get accused of targeting a user if they're constantly getting posts deleted, even if its warranted.

I'd say its toxic in terms of the mindset and what it can do to the site.

Practically we treat it as we would treat low quality content, mass vandalism, or any other appropriate guidelines that fit.

4
  • If what you write is true, then there should not be a delete button on net negative answers. That can't be right. I'm sure there are further cases than those two you listed. Commented Oct 16 at 12:48
  • 9
    I categorically disagree, mostly because the system seems to disagree too. You are rewarded with a restoration of reputation if you remove a new-enough negatively-scored answer of yours. If our goal on paper is "build a library", and users signal to you that your contribution to the library is bad, then I argue that self-removal not only makes sense, but is good, actually, absent other improvements from the author (which I'd certainly say are preferable to deletion, though). Without direct guidance or incentives to the contrary, "toxic" seems an exceedingly unfitting label to me. Commented Oct 16 at 16:32
  • I don't understand the argument against high-rep users deleting downvoted posts (skipping out on the downsides of low-quality posting). I thought the goal of this site was to create a repository of knowledge, not to punish people. If an answer is wrong (which downvoted answers usually are), forcing users to keep it there and accumulate more downvotes only serves as a punishment, and it does nothing to improve the quality of the site. Commented Nov 22 at 14:09
  • I'd like high rep users to avoid throwing things at a wall and see what sticks, but rather post quality content - so its less about deletion than posting things worth deleting Commented Nov 22 at 14:54
7

You ask specifically for policy, but I'm going to approach this from a more systemic standpoint, particularly concerning downvotes– based on how the site works, no, it's not broadly discouraged, and would fall more into a site norms area. If anything, I'd argue that current site design actively encourages this behavior through its incentive structures.

As it stands:

Particularly with the latter, just from an incentives standpoint it seems crystal clear why people would want to do this: they tangibly benefit! The more critical the voting feedback, the more reason to wipe the slate clean.

But zooming out a bit, if we view the goal of Stack Exchange sites as "building a library" (or many), then if I post something in earnest that other users signal as a bad contribution, then self-removal arguably isn't just sensical for me, but actively beneficial to the library! I don't think it's unreasonable to assert that we don't benefit from "bad" answers, in the general sense.

Given, I'd also say that fixing is superior to self-censorship. But depending on the site and the context, that may range from easy to downright impossible, especially on sites scoped to a less objective subject matter than, say, programming, or if my answer turns out to be just incorrect.

In practice, there's nuance here, and things are downvoted and deleted for a myriad of reasons– I'm not privy to the specific case referenced in the question that produced the quoted comment, and I can definitely imagine a hypothetical where a user deletes their own post in the midst of otherwise acting poorly.

But overall, I just find it really difficult to see this as a bad behavior when I look at how the site currently functions and rewards users; much less call it "toxic" for a user to remove something that the community around them has explicitly signaled as poor, low quality, and/ or low utility.

With no apparent incentives or guidance to the contrary, it feels like a humongous ask to expect anyone, old or new, high or low rep, to want to weather downvotes when they're given little to no reason to do so.

It's a valid caveat, however, that "negatively received" may not necessarily equate to "negatively scored". If, for example, a user is unable to take genuine, constructive criticism, or they rollback helpful edits, or delete their post after meaningful refinement by the community– that feels more legitimately classifiable as "toxic" to me. Collaboration is the name of the game here; it's woven into the DNA of SE sites, and it's not something a user can opt-out of.


Some other thoughts that came to mind that didn't weave well into the above:

  • The asymmetric reputation impact of votes (+10/-2), I think, does reward salvaging answers over deleting them, since it only takes a single upvote to nullify the reputation impact of five downvotes
  • I'm definitely not saying that the current design can't or doesn't necessarily have negative side/ knock-on effects
  • Whether site design should be this way, or whether it could be better if tweaked to some degree, is a worthwhile, but separate, discussion

This post is an expansion of a comment I made under @JourneymanGeek's answer.

1
  • 3
    As Glorfindel mentions, there is the risk of an answer ban, especially for low rep members, although (I think) answer bans are less likely than question bans. But for both deleted questions & snswers there's the problem that deleted posts without a positive score are counted by the ban algorithm. That's a bit counter-intuitive to many people, especially newbies, who think deleting destroys the badness. And once you delete a post you're likely to forget about it, and unlikely to fix it, especially if you don't realise that it's still having a negative effect. Commented Oct 17 at 1:54
3

I've seen comments on Super User Meta where users call it "toxic" or "bad behavior" to post an answer, and then delete it when it receives negative feedback or downvotes.

Since we are talking about my comment. I find it extremely annoying to spend 5 minutes of research, and subsequently submit feedback to a user who submitted a factual inaccurate answer, only to find it instantly deleted the moment the feedback was provided while simultaneously constantly complaining about the downvotes they received to past answers without receiving that very feedback that was provided which ended up causing their answer to be instantly deleted.

So I absolutely find the behavior i just described extremely toxic. I am not talking about the rare case of as a question answerer, not understanding that rare question, and completely missing the mark with an answer. I am talking about deleting any answer that receives any negative feedback or downvotes, nearly instantly, consistently (as example at least once a day).

What I describe absolutely is routine with an extremely small subset of users, and they have to date, never been impacted by an answer ban because of the thousands of answers that they submitted.

I have a problem if I spend time researching a topic before submitting an answer or researching if an answer that was already submitted is actually accurate, only to find that contribution that was “deemed good enough to submit” being deleted, once any negative feedback is provided publicly. It shows me that perhaps the negative feedback is not worth sharing, if it means, the contribution will simply be deleted after a single comment.

Heck, what do I know, I am only human.

2
  • 1
    I don’t know the full backstory, but your case seems like an exception and probably shouldn’t be taken as a general example. I'll update my post accordingly. Commented Oct 23 at 8:28
  • 1
    @Velvet - I didn’t provide the full backstory for a very good reason, is that the specifics don’t matter, but it’s more common then you realize. Commented Oct 23 at 12:32
3

Deleting a negatively received answer is by no means "toxic" behavior, by any definition of the word "toxic". It is usually a good thing, because most of the time downvoted answers are wrong, and the goal of the Stack Exchange network is to become a repository of knowledge (of true, high-quality knowledge, of course).

Some argue that deleting own downvoted posts unfairly avoids loss of reputation. But in this case, that is a good thing because it removes wrong or low-quality content from the site. In any case, calling it "toxic" is really a stretch.

Deleting poorly received questions is a different thing, because someone may have spent effort answering it and that answer will also get deleted with the question. But the site does not lose anything of value if a low-quality answer is deleted by its poster.

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.