7
$\begingroup$

It is decided[1] that deleting answers just because they're wrong/incorrect is not condoned in StackExchange-land.

However, I feel it is a different situation when the answer is a borderline low-quality answer. What do I mean by "borderline low-quality?" I think a representative definition would be:

Borderline low-quality answers would be perfectly acceptable if they were correct, yet are close enough to "low-quality" that they could warrant deletion when they're wrong. These answers are typically short, don't give much reasoning, and don't really add value when they're wrong.

(I understand wrong answers can be informative, but the ones I'm discussing are not.)

As a concrete example, please see this review I just performed. This answer doesn't show enough work or demonstrate a process that would redeem it.

However, it's not clearly low quality--after all, if the answer was correct, it would be a perfectly acceptable answer. Hence, I consider it "borderline low-quality." I recommended deletion, but now I'm wondering what the community thinks about responding to these answers.

tl;dr: Should we delete borderline low-quality, wrong answers?

[1] Granted, nothing is ever "decided," community opinions change, etc. I'm writing this question under the assumption that this really was decided. It would be a perfectly fine answer (IMO) to say "Math.SE is going to forget about that, and delete wrong answers." Either way, my question is answered.

$\endgroup$
5
  • 1
    $\begingroup$ How is this question different from What is “very low quality”? $\endgroup$
    – user147263
    Commented Jul 16, 2014 at 1:39
  • $\begingroup$ @Thisismuchhealthier. "What is 'very low quality'" is asking for a definition of very low-quality answers. Furthermore, it is asking "when should I flag a question/answer?" My question is essentially asking, "can the quality of an answer vary depending on whether the answer is right or wrong?" $\endgroup$
    – apnorton
    Commented Jul 16, 2014 at 1:44
  • 1
    $\begingroup$ Okay, maybe it is different enough so that new opinions will be added. For me, the question is answered by the flowchart I posted there. In this example, the answer is an attempt to answer the question, and does not have any formatting issues. As you correctly observed, an answer of the form: "counterexample: $f'(x)= \dots$" can be a perfectly good answer. This one falls short only because of being mathematically wrong. $\endgroup$
    – user147263
    Commented Jul 16, 2014 at 1:46
  • 1
    $\begingroup$ That was the dilemma I faced, and I spent some time wrestling with it before casting the final "leave open" as it turned out. I did downvote the Answer, and I left a Comment detailing how thoroughly wrong the Answer was. Final decision point was that the very wrongness of the Answer seemed to have educational value. Of course the morning-after I have a bit of regret... $\endgroup$
    – hardmath
    Commented Jul 16, 2014 at 11:47
  • 3
    $\begingroup$ @anorton The answer is now deleted. I don't think that the MSO link shows that the matter "is decided" (even on SO). Many wrong answers with no pedagogical value have been deleted here. While some wrong answers are helpful because they serve as warnings to help avoid common pitfalls, they are relatively rare (I don't recall any here off the top of my head, but probably some do exist). $\endgroup$
    – Bill Dubuque
    Commented Jul 16, 2014 at 13:49

1 Answer 1

Reset to default
0
$\begingroup$

This idea makes sense. Answers that are "wrong" and "borderline low quality" have two strikes (or at least one and a half) against them.

An otherwise "low quality" answer that's "right" has a redeeming feature. An otherwise "wrong" answer that goes into some detail, explanation, etc., might be salvaged (and made into a pretty good answer) by a tweak or two.

But an answer that has all negative and no positive qualities has "NO redeeeming feature."

$\endgroup$
1
  • 2
    $\begingroup$ If you have nothing agaisn answers that could be redeemed in one step, why raise brows on answers that could be redeemed in two steps? $\endgroup$
    – Vladimir
    Commented Jul 27, 2014 at 19:51

You must log in to answer this question.

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