Recently, I came across this answer to an old question, which attempts to prove Fermat's Last Theorem (edit: the answer has now been deleted). It takes almost no effort to see why the argument there is wrong, but somehow, the answer has one upvote and no downvotes. The answer states verbatim:

My investigation of FLT in the last year or so would indicate that there is a simple solution, but you'll need to accept the above constraints. I understand that Wiles' et al solved FLT with out taking advantage of those constraints and made the problem much much harder than it actually is (IMHO).

While this meta post serves to draw attention to this answer in particular, I also want to open a discussion about what we should do when seeing such answers. Unlike questions, which have a better chance of being seen by more people and hence taken down as off-topic, answers seem to be able to slip through the cracks sometimes. In case we see something like this, should we flag for moderator intervention, stating that the premise of the answer is blatantly false? Or should we just downvote and hope that enough people see it (and downvote) to invalidate the answer? Or should we do something else?

  • 9
    $\begingroup$ Downvoted. It might be better to advertise this in C.R.U.D.E.. There are a lot of users concerned about the quality of the material on the site there. I don't think getting it deleted will take long :-) Just cast the second vote to delete. $\endgroup$ – Jyrki Lahtonen Feb 3 at 15:31
  • 2
  • 3
    $\begingroup$ And some past discussions related to actions of users on incorrect answers: Don't flag wrong answers?, How to flag kook answers? and Should we vote to delete wrong answers? $\endgroup$ – Martin Sleziak Feb 3 at 16:05
  • 13
    $\begingroup$ If it's clearly wrong, at least downvote and comment. Then at least future readers will see the comment. If it's a simple slip like inadvertently answering the wrong question or mistyping an expression, the questioner may well hurriedly correct it or delete it. But if they don't, readers will realise there's a problem. $\endgroup$ – timtfj Feb 3 at 19:49
  • 1
    $\begingroup$ I think the one upvote is most likely from the questioner in these situations, though I've a feeling it's also possible to upvote your own answer, which I suppose someone claiming to have done the impossible might do. $\endgroup$ – timtfj Feb 3 at 19:54
  • 4
    $\begingroup$ "though I've a feeling it's also possible to upvote your own answer" -- It isn't. Unless you want to suppose shenanigans about people having multiple accounts; I'm not sure how SE handles such events. $\endgroup$ – Eevee Trainer Feb 4 at 2:08
  • 5
    $\begingroup$ If you need more votes on bogus questions/answers, post it in the CRUDE chat-room as Jryki mentioned, and you're welcome to ping me. I think most regulars there won't mind you pinging them as well, for particularly difficult kook posts (which need multiple downvotes before it can even be deleted). You're actually quite lucky here; some kook answers need >10 downvotes. $\endgroup$ – user21820 Feb 4 at 7:40
  • 1
    $\begingroup$ @EeveeTrainer I'm probably mixing it up with accepting your own answer (which the help centre says you can do but won't get any reputation points for). $\endgroup$ – timtfj Feb 4 at 13:03
  • $\begingroup$ Additionallly I'd try and look out a bit whether there is some indication of being a math-bot answer (may be in trainingphase of the bot) - don't really know why at this example this comes into my mind ... $\endgroup$ – Gottfried Helms Feb 8 at 11:22
  • $\begingroup$ My issue I have had in the past is that someone will post an answer to a problem which is accepted and gets 10 upvotes, I explain to the person who posted the answer that their answer is manifestly wrong and can't be right and that they must do something to fix it. They just say something like, it must be right as I can't see what to do otherwise, and then the answer stays there to confuse potentially hundreds of other people. $\endgroup$ – Tom Feb 8 at 15:00
  • $\begingroup$ @Tom can you give an example please? I honestly haven't seen a question with an answer which is mathematically wrong, yet obtain 10 upvotes and gets accepted. Usually, even the slightest mathematical mistakes will be noticed and the answer is either ignored (0 up/down votes) or down voted into oblivion. $\endgroup$ – YiFan Feb 8 at 15:58
  • $\begingroup$ OK so I was exaggerating but this answer here was accepted and had 4 upvotes, I explained again and again that the answer could not be correct but nothing was done about it and the answer was left there as the accepted answer math.stackexchange.com/questions/2704071/… $\endgroup$ – Tom Feb 8 at 19:04
  • 3
    $\begingroup$ @timtfj: Commenting on badly wrong answers is nice and all, but sometimes such posters are, shall we say, not open to criticism, and it can lead to an unpleasant interaction. I once posted what I thought was a quite polite comment on an incorrect and rather confused answer, and the poster spent the next few hours creating new accounts via Tor to post questions with titles like "I f***ed Nate Eldredge's mom". Sent me 50 or 60 such emails as well. So it's nice to leave comments, but if it feels like someone you don't want to engage with, I think silent downvotes are fine too. $\endgroup$ – Nate Eldredge Feb 9 at 2:28
  • $\begingroup$ @Nate Ouch. That's pretty extreme. I can see why silence would be the best option there. It sounds horrendous. $\endgroup$ – timtfj Feb 9 at 12:47
  • $\begingroup$ I've down-voted the answer that I discussed in my post so it's now down to 2, that's an improvement, but it's still on there as the accepted answer with no alternatives. $\endgroup$ – Tom Feb 9 at 14:08

I am fairly new on stackexchange and I don't have enough reputation to comment. So, I am compelled to write an answer, although my first choice is to put a comment under your question. But now, that I have enough legroom, I will try to answer it in detail according to my understanding.

My Suggestion: If you (we) come across such answer(s) which are blatantly wrong then you should either provide a better explanation to why that answer is utterly invalid and/or set on wrong premise(s). Sometimes, we infer that something is wrong, but can't provide better expression to our thoughts to write a nice counter comment (or answer), at that time we can downvote and bring answer to the attention of the moderator.

You can also bring this answer to the attention of other active members and ask of their view. If they agree then you can request them to downvote and report-to-mod at same time.

Lastly, you can open up a chat with the answerer and debate to why they think it is the right answer. During the process when you elicit more information or explanation then there are chances that they get aware to why their assertion is entirely wrong. Majority of the time they will understand.

After all, no one earns money or benefit otherwise here on stackexchange. Everyone is trying to help a person who comes here with some inquiry/concern/question/query.

  • 5
    $\begingroup$ Sure, thanks for the response. But I was less thinking along the lines of "controversial, in my opinion wrong but others may disagree" posts, but "ridiculously wrong and absurdly misleading". In such cases I don't think it's appropriate to open a long winded discussion as you suggest, but I think a rapid deletion would be more in order. Also, welcome to M.SE! $\endgroup$ – YiFan Feb 5 at 9:34
  • $\begingroup$ @EVG Also, see Nate Eldredge's comment about engaging with someone it would have been better to leave alone! That hadn't really occurred to me as likely to happen on here. $\endgroup$ – timtfj Feb 9 at 12:54
  • $\begingroup$ @timtfj I read that comment, but I am not experience enough on this site. Up till now all the questions that I have asked on this site have been answered wisely and with utmost humble attitude! $\endgroup$ – EVG Feb 9 at 17:08
  • $\begingroup$ @YiFan I agree with you. I guess, moderator can be of great help. Few years back I was bit active on English.Stackexchange and moderator there were very strict. I am not experienced enough to talk about math.stackexchange, but I am pretty sure majority of answerer are well mannered and well qualified to answer. $\endgroup$ – EVG Feb 9 at 17:16

You must log in to answer this question.

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