15
$\begingroup$

Apologies if this should go in the "requests to undelete questions" thread but this seems to be technically different.

I noticed that this answer https://mathoverflow.net/a/465960/ has recently been deleted by a moderator. Would it be possible to say why? I don't immediately know if the answer is a good one but I don't see why it would be immediately judged to be of low quality or in bad faith.

$\endgroup$
4
  • 4
    $\begingroup$ Perhaps it is suspected to be of chat-gpt origin. I don't have the expertise to justify having an opinion on the matter, it just reads gpt-ish to me. $\endgroup$
    – Gerry Myerson
    Commented Feb 27 at 2:55
  • 10
    $\begingroup$ @GerryMyerson I am a functional analyst, Michael Frank is a known researcher on Hilbert modules ( imn.htwk-leipzig.de/~mfrank/hilmod.html ), and that answer contains several correct statements that relate to the first comment on the original post. $\endgroup$
    – Yemon Choi
    Commented Feb 27 at 4:12
  • 1
    $\begingroup$ Ah. My apologies to Michael Frank. $\endgroup$
    – Gerry Myerson
    Commented Feb 27 at 5:02
  • 4
    $\begingroup$ I've added a line to that answer to clarify that it is an extended comment. Without this line I perfectly understand that it was misread as a IA-answer (although I hadn't flagged it). $\endgroup$
    – YCor
    Commented Mar 4 at 8:11

1 Answer 1

Reset to default
21
$\begingroup$

My apologies. — The reasons why I mistakenly deleted this answer were the following:

  • The answer got flagged as "not an answer" by a user who usually raises helpful flags.

  • On a first glance, it seemed plausible that the answer was written by ChatGPT.

  • The answer came more than one year after the question was asked and got an accepted answer.

  • The answer was posted as the first post of an unregistered user.

$\endgroup$
8
  • 13
    $\begingroup$ Given the circumstances, you might think about sending a personal email to the author of the post explaining the situation and apologizing. From what Yemon Choi said, it sounds like they are exactly the kind of user we want to attract and not drive away. $\endgroup$
    – Andy Putman
    Commented Feb 27 at 12:21
  • 14
    $\begingroup$ @AndyPutman Done. $\endgroup$
    – Stefan Kohl Mod
    Commented Feb 27 at 14:24
  • 8
    $\begingroup$ In Stefan's defense, accounts posting AI-generated semi-nonsense seem to have sprung up like mushrooms lately (if a non-moderator can already notice...). $\endgroup$
    – Christian Remling
    Commented Feb 27 at 15:08
  • 12
    $\begingroup$ Thanks for the transparency here. If anything, the fact that the prevalence of AI junk answers sometimes leads to moderators deleting good, human answers is just another annoying facet of the rise of this AI junk. $\endgroup$
    – Sam Hopkins
    Commented Feb 27 at 22:29
  • 6
    $\begingroup$ Thanks Stefan for the prompt and open response. Hopefully all's well that ends well. $\endgroup$
    – Yemon Choi
    Commented Feb 27 at 23:22
  • 2
    $\begingroup$ @YemonChoi: I was the one who has raised the NaN flag, simply because the post under discussion does not even attempt to answer the question. It is only a long comment to a side discussion. It seems intended to answer the question that Matthew Daws' has raised in his comments about what "finitely generated" means. Upvoting the post under discussion (12 upvotes!) just to apologize to its author for the previous deletion is a wrong way to use the voting mechanism: Michael Frank's non-answer now has a bigger score than the accepted answer itself (7), which is ridiculous. $\endgroup$
    – Alex M.
    Commented Mar 5 at 7:54
  • 1
    $\begingroup$ And thank you, Stefan Kohl, for complimenting the "usual helpfulness" of my flags! $\endgroup$
    – Alex M.
    Commented Mar 5 at 7:54
  • 2
    $\begingroup$ I disagree that Michael Frank's post is not an answer. The question asker themselves pointed out that they were confused about or didn't know the precise terminology that went into the result they were interested in. So clarification about this terminology qualifies as at least a partial answer in my view. I do not believe a post has to completely and definitively answer a question to qualify as an answer. $\endgroup$
    – Sam Hopkins
    Commented Mar 6 at 1:56

You must log in to answer this question.

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