When you click on the "envelope" next to your name on the top of the page, you get a summary of the recent changes. Clicking on the "Reputation" tab will tell you the up and down votes on your questions, your own downvotes, and any acceptance of your answers, as they reflect changes in your reputation.

Just now, I noticed that my rep had gone down by 15 points, presumably because an answer that had been previously marked as "accepted" was "dis-accepted." I would like to find out which one it was (and to see if a different answer was accepted instead). But it seems that this is not displayed in that page.

Am I correct that such changes would not be displayed? Presumably, the +15 tag with the acceptance has disappeared from whichever day/week/month summary it was on before, but I unless I happen to remember which answers had been accepted and compare it to the current status of my answers, it seems difficult to figure it out.

If there is a way in which this is displayed, what is it? If there is no way, is there any way that information about such "dis-acceptance" changes might also be displayed in the summary list?


4 Answers 4


There's a new discussion on meta.SO, started by Nick Craver (a developer): How do you want to be notified of unaccepted answers? It appears they're working on completing the feature-requests mentioned in Jonas' answer.

Update: This has been completed now; as of 2011-09-22, users can see their unaccepted answers in the reputation report.

  • $\begingroup$ Thank you for the heads up! $\endgroup$ Sep 7, 2011 at 15:47

Update Sep. 22 23, 2011: The answer below is outdated. See Hendrik Vogt's answer for the news. (I'm adding this because this answer is currently accepted and might be misleading without a disclaimer.)

See these meta.stackoverflow.com posts:

Notification when my answer is unaccepted?

Reputation notification of deselected answer?

Please show us when we lose an accepted answer

They are all tagged "feature-request" and the earliest one is highly upvoted, but I don't see any feedback from the administrators. One of the accepted answers reads:

I think it makes sense that you don't get notification because the deselection is actually wiping out the vote record in the DB.

  • $\begingroup$ Okay, I can see that; it seems, though, that it should be possible that the same script that wipes the DB could place a notification on your activity-page of some kind. $\endgroup$ Dec 20, 2010 at 1:05
  • 2
    $\begingroup$ @Arturo: I agree, and many meta.SO voters seem to agree that it would be a good idea to do so. $\endgroup$ Dec 20, 2010 at 1:07
  • 5
    $\begingroup$ I guess we should encourage the "unaccepter" to leave a comment. I suppose it is the polite thing to do too... $\endgroup$
    – Aryabhata
    Dec 20, 2010 at 3:30

I just had this happen to me and I would have liked to be notified about it. Not out of a desire to be the one with the accepted answer, or to enter in to a shoot-out over reputation, but to maximise the chance that the OP gets an answer to their question.

My answer was the only answer to that question. I cited a theorem that was close to the result being asked for. The OP accepted it, but then (I'm guessing) de-accepted it due to some comments about how close was close enough. Fortunately, I saw those comments and realised that there was still a gap between the result that I'd cited and the result that the OP wanted. Even more fortunately, the gap was something that I'd proven in one of my own papers. So I was able to modify my answer to better answer the OP's question.

However, to even see this, I relied on comments to my answer. Being notified about the de-acceptance would be a much more reliable method of finding out that an answer I'd posted hadn't actually answered the OP's question given that it had seemed, at one point, that it had.

So I'm off now to meta.SO to vote for those feature requests.

(The question in question is: If $g^{-1} \circ f \circ g$ is $C^\infty$ whenever $f$ is $C^\infty$, must $g$ be $C^\infty$?)


Answered status and especially "which answer is, in the OP's opinion, most acceptance-worthy" are poster-dependent metadata. They are about the relationship between poster and answer, and external to the relationship between question and answer(s).

Given that distinction, I think it is against the impersonal Q&A style of the site (see earlier thread "SE is not a social networking site") to create additional connections between the OP metadata and other users, in this case notifying the authors of the old and new accepted answers, as though they have acquired a piece of the question entitling them to status updates. It creates bogus connotations of ownership, rejection and superiority (of answers), and there are already users who fight over which answer truly satisfies the OP's intention, rather than posting answers for whatever value those may have to the user population and/or the OP.

Accentuating the acceptance status of answers leads to comment wars over which answer is correct, when the acceptance itself is often nothing more than a personal opinion of a judge who is -- by definition! -- unqualified.

  • 6
    $\begingroup$ I have always worried that "de-acceptance notifications" might lead to excessive angst. However, if the argument is "I want to see how I can make my answer even better to compete with the new accepted answer", then perhaps that's a net positive? $\endgroup$ Dec 20, 2010 at 7:21
  • 1
    $\begingroup$ @Jeff: in the ideal scenario of positive answer competition without comment wars or other friction, wouldn't different answers tend to converge (like competitors in an economy or ecosystem), borrowing material from each other or getting closer as they expand their territory in "answer space"? I guess it can be good to harmonize different approaches or have redundancy, but if answers get longer only to end up saying more of the same things it can lose some of the individuality and pithiness. $\endgroup$
    – T..
    Dec 20, 2010 at 7:36
  • $\begingroup$ @T.. well Jon Skeet, one of the top SO users, argued that un-accept notifications would let him determine how to fix his un-accepted answer if it was inaccurate in some way. However I am not sure Jon's exemplary behavior is necessarily representative of the average user.. $\endgroup$ Dec 20, 2010 at 9:00
  • 1
    $\begingroup$ @Jeff: an abstraction of Skeet's comment is that any way of conveying additional information about answers will generally improve them, sometimes including relation or ranking between answers (or a "this is wrong!" flag reported privately to authors but not moderators). I agree with that version but it is not clear why OP opinion about answers is a valuable information channel or one that should be broadened. Its personal (and often unqualified) nature is an argument for reducing the relative importance of OP-specific feedback in favor of the OP-independent answer-specific type. $\endgroup$
    – T..
    Dec 20, 2010 at 9:25
  • 2
    $\begingroup$ @T.. I agree with you on this, for sure. I'm just offering the only justifiable defense of this feature request I've seen, but I doubt the benefits outweigh the risk for many of the same reasons you outlined above. $\endgroup$ Dec 20, 2010 at 10:05
  • 4
    $\begingroup$ @Jeff: 1. While I can certainly appreciate your concerns about "excessive angst", I think this should be weighed against the positive aspects of such unacceptance notifications. First, such notifications give the answerer feedback that may prove beneficial towards improving future answers. Second, they may serve to help further educate the person whose answer was unaccepted by explicitly pointing out the existence of answer by someone more expert on the topic. (continued) $\endgroup$ Dec 20, 2010 at 16:12
  • 3
    $\begingroup$ @Jeff: 2. (continued from 1). Further, being able to compare which answers are more comprehensible helps one to infer the knowledge level of the questioner - which may be of great help in constructing answers to future questions by the questioner. Far too often such contextual background information is omitted and it is difficult if not impossible to give optimal answers without such information. $\endgroup$ Dec 20, 2010 at 16:19

You must log in to answer this question.

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