20

I just ran into a question which has an answer that is:

  • 100% canonically wrong (as in, its conclusion is directly fully contradicted by canon)

  • But it's old, has 25 upvotes, and is accepted

    (and the user doesn't seem too active so a chance of him accepting a more correct answer seems unlikely).

What avenues do I have to fix this deplorable situation, other than appealing to the better nature of a user who posted an incorrect answer via a comment?

  • 12
    Also, of course I can't tell people what to answer, but please consider NOT posting the old boring "Let SE mechanics work its magic, and your correct answer will eventually get more upvotes than incorrect one". That happens in less than 10% of the cases I observed, especially on 4 year old questions. – DVK-on-Ahch-To Dec 17 '15 at 17:29
  • 1
    not a dupe imo. voting to reopen – Himarm Dec 17 '15 at 18:00
  • 5
    The situation with an extremly old question with very high votes js a big problem especially of the question asker is no longer active, people will search and find a wrong answer forever – Himarm Dec 17 '15 at 18:06
  • @Himarm If one were about preventing such answers, and one were about post-treatment, I'd say reopen, but they're both about dealing with general cases of upvoted, incorrect answers after-the-fact. – user31178 Dec 17 '15 at 18:49
  • What's the question/answer? There's nothing at all we can do about acceptance, but we could at least let the meta effect take care of the score – Jason Baker Dec 17 '15 at 18:50
  • 2
    @CreationEdge the old question was about a upvote issue in a couple day old question, supposedly this new question is about dealing with a high upvoted wrong answer, YEARS old, which make it a different topic imo, patience works for a couple day old question, but as we all know year old questions are another mater. – Himarm Dec 17 '15 at 19:02
  • 1
    @CreationEdge - also, original question didn't have the added issue of the wrong question already having been accepted years ago. – DVK-on-Ahch-To Dec 17 '15 at 19:03
  • Minor update: the OP of the question I was thinking of was actually gracious enough to change his acceptance. However, this Meta question is generic and therefore still stands as-is. – DVK-on-Ahch-To Dec 17 '15 at 19:04
  • @DVK It's not generic if it's explicitly referring to a question with a +25 accepted wrong answer with an inactive OP. That's pretty specific! But I can see the difference on the other question is being more about wrong answers being given, rather than wrong answers being accepted. – user31178 Dec 17 '15 at 19:08
  • @CreationEdge - it is generic because there are other such questions, I'm asking how to deal with them. And because I chose not to link the question, though one can probably ferret out which one it is simply because I bothered posting the correct asnwer recently - but NOT otherwise. – DVK-on-Ahch-To Dec 17 '15 at 19:13
  • 3
    It's incredibly important to remember that the fact that an answer is "accepted" never means that it is "right", just that it was deemed as the most helpful by the asker. – phantom42 Dec 17 '15 at 21:00
  • 3
    Isn't this exactly what the bounties are for? To get a authoritative/updated answer? – phantom42 Dec 17 '15 at 21:02
  • 4
    @phantom42 - that works on SO (because even wrong answer can help you solve a programming problem by giving you ideas and approaches)... but how can an answer which is 100% wrong be "helpful" to an asker on SFF? – DVK-on-Ahch-To Dec 17 '15 at 21:02
  • 3
    BTW: I agree that this is an issue, I'm just not sure the current procedures in place aren't enough. – phantom42 Dec 17 '15 at 21:04
  • 1
    This brings up the larger question of, "Should canon be considered more important than popular opinion?" Personally, I care nothing for "canon", especially when "canon" is decided by some corporation (like Disney in the case of Star Wars). If I ever do ask a question on this site, the answer I accept will be the one that seems most consistent with the source material that I consider to be important, and my decision to accept will have nothing to do with the any corporation says, and it will only be influenced (not determined) by the thoughts of the creator(s) of the work in question. – Todd Wilcox Dec 21 '15 at 21:49
9

In my opinion a new question should be asked, with a correct answer, and the Old question, with the Incorrect answer should be tagged as the dupe of the NEW question.

this seems to have been done before https://scifi.meta.stackexchange.com/a/6601/29220

as alex says

Just because two questions are duplicates doesn't mean we have to close the new question as a dupe of the old one. We make the question with the best answers the "canonical" question, and closes others as a dupe of that, regardless of the order in which they were posted.

again common consensus seems to be, https://scifi.meta.stackexchange.com/a/4746/29220

Izkata says,

No. In general, choose the better one to keep open, as long as it is significantly better.

Beofett argues,

If they are essentially the same, the one with the better answers should be left open, and the other closed as a duplicate. This usually, but not always, results in the older version being left open.

  • In the instance you've highlighted, the older question didn't have an answer. – Valorum Dec 18 '15 at 9:01
  • 3
    Your contention of which answer is "better" seems highly subjective. Others may not feel the same way. "Canonically valid" is not the same as "best answer". – Valorum Dec 18 '15 at 9:02
  • 2
    I agree with Richard and want to go even further: who decides what the "correct" answer is? Currently the system is that the asker of the question decides what answer to accept, by whatever criteria they deem important, and other members vote on answers, likewise based on their own criteria. I don't think any SE site has claimed that the system is designed to arrive at any kind of objective "truth", rather that "good" answers rise to the top. "Good" is subjective. – Todd Wilcox Dec 21 '15 at 21:54
  • @ToddWilcox this site is very easy to literally have FACTS as answers. if a user 4 years ago, accepts some answer it found useful but if factually false, we need a way to point out to future readers that while the author of the question liked this answer, its actually literally wrong, and as dvk notes, 4 year old questions get far to little traffic for a high upvoted, accepted, and factually wrong answer, to get downvoted enough, that a new correct answer catches readers eye. – Himarm Dec 21 '15 at 22:08
  • 1
    @Himarm I could see that being a serious problem on Server Fault or Aviation, but here it's hard to believe that there are actually in-universe hard and fast "facts" about science fiction or fantasy. I mean, "fiction" and "fantasy" - it's in the name! Even direct quotes or images from source material are open to interpretation. – Todd Wilcox Dec 21 '15 at 22:16
  • 1
    @ToddWilcox alot of questions are answered by word of god (interviews from authors) or from further books, example question about why this happened in harrypotter book 4, writing an answer before book 5 is out, may become obsolete, upon reading 5, but now we have a year old question and answer that are "wrong" and have to attempt to re-answer it, and hope that the user with 100 rep who asked the question last year didnt quit stack completely. obviously they were many subjective question and answers, but we have ALOT of very distinct factual answers. – Himarm Dec 21 '15 at 22:18
  • 1
    @Himarm You and I don't even have the same opinion on whether it is meaningful to consider anything in-universe to be a fact or whether there is one objectively right answer to any question on this site. We probably also disagree on the very nature of truth itself. I doubt we will come to agreement. You and DVK might do well to note that "many of the truths we cling to depend greatly on our point of view". – Todd Wilcox Dec 21 '15 at 22:22
  • 2
    Of course this entails a race condition — you need to make sure that the old question gets closed as a dupe of the new question before the new question gets closed as a dupe of the old question, which is likely to happen faster. – Peregrine Rook Dec 26 '15 at 7:50
  • @PeregrineRook - Creating a straight dupe, with the sole intention of promoting a newer answer would seem to be an abuse of the system. A moderator should step in and merge the questions. – Valorum Dec 28 '15 at 9:32
  • @Richard: Why, exactly, are you addressing that comment to me?  Himarm is the one who said, “In my opinion a new question should be asked … and the Old question … should be tagged as the dupe of the NEW question.”  All I did was make an observation regarding the logistics of Himarm’s stated opinion.   And, if you disagree with Himarm’s opinion, perhaps you should notify alexwlchan and SQB, Null, Keen♦, whose statements and actions support it.  … … … … … … … … …  P.S. What does “on strike” mean? – Peregrine Rook Dec 28 '15 at 20:58
  • @PeregrineRook - I was addressing the logistics, hence why I addressed it to you. As to my "strike", I've decided to withdraw from editing & flagging in protest at my unwarranted chat ban. – Valorum Dec 28 '15 at 21:08
6

Per my answer to a similar question,

The correct procedure is to add a comment explaining why the answer is wrong. The answer will then be (gradually) outvoted by your correct answer. It might take years but them's is the downside of a vote-based system.

You might also want to address a note to the poster explaining why their answer is wrong and encouraging them to delete or amend their previous answer with the correct canon information.

Notably, this approach has proven successful on more than one occasion. In this instance (the one that prompted the question in the first place), the OP has actually changed their acceptance:

DVK: "Minor update: the OP of the question I was thinking of was actually gracious enough to change his acceptance. However, this Meta question is generic and therefore still stands as-is."

On other questions, I've personally been successful in persuading both the poster of a superseded answer and the question's original OP to add notes and change their acceptance, resulting in my new answer being upvoted ahead of the old answer:

How is canonicity of derivative works determined for Star Wars?

You must log in to answer this question.

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