I just had this experience and was wondering if there is anything to be done about it:

• I see a question on the front page.
• Have (what I think is) a nice approach to it so spend 5-10 minutes writing it up
• Hit "Save" and be notified that the question has been deleted.
• Arrrgh!

The question in question seems perfectly fine and appropriate for Math.SE, but I am a long way from having a high enough reputation to undelete. My questions/issues are:

• I think this is bad site etiquette. Is that accepted or am I way off?
• Does the poster (who seems a bit new here, 42 days) get any push back that this is a bit rude?
• I think I'm a solid mid-range participant here. I like reading other people's answers, and on the occasions when I do see one where I've got a good response I like sharing and working with the questioner. Things like this really make me less likely to go to the trouble of writing up answers. This might just be me, or it might not happen that often, but I do think it's bad for the site.
• It is not unknown for a user to post a question, get an answer, and then delete the question. We can speculate that the user was trying to cover his tracks, to hide from other parties the help he got from this site. This behavior has been discussed here before and is considered to be a bad thing. There is considerable sentiment favoring undeletion of the question in such circumstances. There is a thread here on meta devoted to requests for reopening and/or undeletion of questions. My apologies for being unable to include specific links (but see the Related questions). – Gerry Myerson Feb 15 at 21:30
• By the way, I'd be surprised if the question on main is not a duplicate, indeed, I suspect it has been asked and answered on main several times over the last ten years. Even if it is undeleted, it might wind up getting closed and deleted as a duplicate. – Gerry Myerson Feb 15 at 21:35
• And... undeleted. It's actually a nice answer. – Arnaud Mortier Feb 15 at 22:06
• @GerryMyerson: Yeah, "closed as a duplicate" is fine. It happens sometimes. Also, the original poster seems to have left up a few other questions at about the same level, so I think they are participating with good intentions. I wonder if they just weren't thinking. And I'll use the "please undelete" thread next time. – JonathanZ supports MonicaC Feb 15 at 22:49
• If a user deletes his question just after receiving an answer, before that answer is able to get an upvote, it can be flagged for mod undeletion. If it's before posting the answer, the undelete thread is best. – Alexander Gruber Feb 16 at 2:26
• @AlexanderGruber: Okay. And just to make it clear, this was a case of the latter type. The poster deleted their question before seeing my answer. I want to stress that because there seems to be some weirdly aggressive commenting happening in my answer, and I have neither evidence or suspicion that the poster was acting in bad faith. I presume they just didn't think through what they were doing and I certainly don't want people to be unpleasant to them. – JonathanZ supports MonicaC Feb 16 at 3:48
• Related: Self-deletion of question while typing answer (and other questions linked there). – Martin Sleziak Feb 18 at 6:43
• For what it's worth, this situation is a fairly rare coincidence; I think it may have happened to me twice out of almost a thousand answers - I wouldn't be too discouraged by it. – Milo Brandt Feb 19 at 2:30
• @Milo Brandt: I read this meta thread yesterday, and today a question was deleted while I was in the middle of typing a comment (but at least not an answer). I was going to say that it might be interesting to investigate the nature of the solutions to $(x-a)=A$ and $(x-a)(x-b)=A$ and $(x-a)(x-b)(x-c)=A$ and $\ldots$ (I guess one approach is to consider vertical translations of the graphs of the left-hand sides assuming, without loss of generality, that $a \leq b \leq c \leq \cdots)$ – Dave L. Renfro Feb 22 at 15:50
• (10 minutes later) Now that I think about it, I suspect (not sure, though) that every algebraic equation with real coefficients can be put into this form for some constants $A,$ $a,$ $b,$ $c,\; \ldots$ – Dave L. Renfro Feb 22 at 16:03