Say $X$ posts an incorrect, or at least incomplete, answer. I post a complete answer. X edits his or her answer - the new version is imo essentially the same as mine, although imo much less well motivated, with the concepts obscured by magical calculations. (For example...)

Even if X doesn't see how the two answers are essentially the same, wouldn't it seem proper for him or her to at least acknowledge that he or she didn't find his or her answer until after reading mine?

Not important, just irritating. If one looks at the time stamps and ignores the edit history it look like X gave the first correct answer, which isn't so.

(Regarding whether X's answer is really the same, or at least derivative: One might note that there was nothing inner-productish in X's first version. More specifically, using the union of the notation in the two answers, the reason his calculation $||f-f_1||=0$ succeeds is that $||f||^2=|\alpha|^2+|\beta|^2$ and $g\perp b_1$, $g\perp b_2$.)

| |
  • 1
    $\begingroup$ How do you definitively know that you deserve credit for how the user improved their question? For all I know, they had their nose in the buffer the entire time and never saw your answer. $\endgroup$ – rschwieb Dec 21 '18 at 17:00
  • 1
    $\begingroup$ There is an exchange of Comments on the Question. I think it might help David to frame this in terms of promoting learning (often a failed approach teaches a deeper lesson than a zipless one). In any case the higher voting for David's Answer will not rub salt in his wound. $\endgroup$ – hardmath Dec 21 '18 at 17:04
  • 3
    $\begingroup$ Anyhow, I think we can all probably agree that wholesale fraud of this sort (like copying the exact same argument from a newer answer and replacing your own) is unacceptable, and we would encourage people to give credit where credit is due. I think your comment on their solution helpfully draws a connection, and that the stakes are pretty small here. $\endgroup$ – rschwieb Dec 21 '18 at 17:04
  • $\begingroup$ @rschwieb The sentence starting "Even if..." indicates that I'm not claiming to know that. After I posted my answer and a comment on his/her first version, before posting his/her second version, he posted a comment stating that yes, I'd given a proof and he hadn't. So regardless of where his/her second version came from, he did know that I'd already posted a correct solutiion before posting his/her second version. $\endgroup$ – David C. Ullrich Dec 21 '18 at 17:14
  • $\begingroup$ @DavidC.Ullrich In that specific case, then no, I don't think people who wind up giving completely independent, ultimately equivalent, but later answers have any obligation to advertise they are late and equivalent. But they could make a gesture if they wanted. I'm still not sure what the proposition of this meta post is, beyond that, though. Is any enforcement being proposed? Or guideline change? It's hard to make headway on issues without having an idea that they are actionable. $\endgroup$ – rschwieb Dec 21 '18 at 17:45
  • 1
    $\begingroup$ @rschwieb The point was just to find out what people thought was appropriate. $\endgroup$ – David C. Ullrich Dec 21 '18 at 17:56
  • $\begingroup$ @DavidC.Ullrich OK: that seems like good enough focus for me! $\endgroup$ – rschwieb Dec 21 '18 at 18:30
  • $\begingroup$ Also related. $\endgroup$ – amWhy Dec 22 '18 at 0:07

The comments on main show that the other answerer does acknowledge your priority. If you wanted this to be more clear, you might have left your first comment on the other answer undeleted.

I think a reader that reads the thread will likely understand the history about correctly: other answer incomplete, your answer complete, the other answer fleshed-out.

Generally the philosophy of the site is that clarity of the presentation of the content takes precedence over high-lighting the historical evolution of the content. Thus, it is alright that the answer does not contain a paragraph explaining the evolution.

I don't think the other answer takes anything away from yours, to the contrary.

Against the more pedestrian approach, the elegance of your approach becomes more clear. In the other direction, for some reader your answer might be more difficult to follow. The question is accessible to everybody with a basic understanding of calculus. Not everybody knows what orthonormal means, or even if, might not be used to using it for functions.

That said, I understand how the situation could look odd to you. You give an answer and right then the other answer gets expanded. But, it seems like a coincidence.

| |
  • $\begingroup$ Your comments on the comments are not so clear to me. In particular, it seemed to me that my comment on the other answer "had" to be deleted, since it no longer made sense as a comment on the revised answer. But never mind that... $\endgroup$ – David C. Ullrich Dec 22 '18 at 14:25
  • $\begingroup$ I do agree that the existence of both answers may well be a good thing, because each of them will be much more clear than the other to a large class of possible readers. $\endgroup$ – David C. Ullrich Dec 22 '18 at 14:27
  • $\begingroup$ Just to give that dead horse one final whack: regarding "seems like a coincidence", see the sentence above starting "Even if...". $\endgroup$ – David C. Ullrich Dec 22 '18 at 14:30
  • 1
    $\begingroup$ Regarding your first comment, yes, in principle, due to the general philosophy of the site I mentioned above. But in practice it'd be more common to have such a comment around than to insist on having the information in the answer post proper. $\endgroup$ – quid Dec 22 '18 at 14:35
  • 1
    $\begingroup$ Regarding the final comment. There is no obligation to do this, but it is not uncommon and certainly considered as good form by many (even though an SE-hardliner might object to such meta-content in an answer post). In the specific case I think there is some form of acknowledgment in the comments, and this seems sufficient to me. $\endgroup$ – quid Dec 22 '18 at 14:39

I've had the experience once or twice of writing an answer, only to discover on improving it that it's simply another version of an existing answer. That happened in Given a book with $100$ pages and a $100$ lemmas, prove that there is some lemma written on the same page as its index where there seem to be two basic approaches, and all the answers use one or the other.

But when I realised what I'd done, I tried to acknowledge that in the answer.

A similar issue is when nobody has posted an answer, but several people have posted comments which together answer the question. A couple of times I've done the job of combining the comments into an answer, but I've set the post as CW in recognition that I've not really contributed an answer, just collated what's already there.

PS apologies for the colossal link—I'm unsure of the syntax for links (and might lose what I've typed if I navigate away to check).

| |
  • $\begingroup$ To use alternate text for a link, put this text inside square brackets with the link inside of round brackets immediately after it. $\endgroup$ – John Omielan Dec 31 '18 at 6:10

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