As a newer user, I am uncertain if the community norms here in CV are different than on Math.SE. There, it appears to be acceptable to ask a user to accept/upvote an answer in cases such as your having posted the only answer and the OP commented that it was helpful (see here). Does that apply here on CV too?

  • 3
    $\begingroup$ Although there are many thoughtful discussions here, no one seems to flag a simple central point of polite behaviour: anything that smacks of asking publicly for praise or reward is at best a little undignified, and if repeated tiresome or even obnoxious. We should all have learned this at an early age. The main qualification is that it's in order to remind or inform people new to the site that question acceptance is a good thing, for them too. That's often best done by someone who didn't write an answer. $\endgroup$ – Nick Cox Jan 19 '17 at 16:07
  • 1
    $\begingroup$ +1 to @NickCox. I often ask inexperienced question askers to consider accepting other people's answers. But I usually hesitate to ask to accept my own ones. (Whenever I feel I have to, my formula is to ask to edit the question to clarify what remains unclear, hinting that if everything is clear then accepting one of the answers is the way to show it.) $\endgroup$ – amoeba says Reinstate Monica Mar 1 '17 at 12:31
  • 1
    $\begingroup$ Yesterday elsewhere I suggested to someone that as they had expressed satisfaction with my answer (the only answer in a thread) they could accept it and we would both gain reputation. The transaction was generous: they tripled their reputation and I didn't. But it's a fine line. If it's someone with reputation in hundreds or thousands or millions, they surely know the game by now and I wouldn't prompt them for acceptance. (This was a niche speciality for which I sometimes appear to be the only really active person answering.) $\endgroup$ – Nick Cox Mar 1 '17 at 12:52

Users failing to respond to good answers by accepting one is annoying, there's no doubt of that. I've seen some wonderful answers go unrecognized by the poster of the question and the lack of response (either accepting or explaining why there are outstanding issues that would stand in the way of it) often strikes me as selfishness in the face of an answerer's generosity. Some regular contributors seem to suffer from this non-response to their efforts shamefully often.

(In my own case, even when I felt that an answer didn't really answer my question after several attempts to convey the issue clearly to them, I have accepted an answer as clear recognition of the answerer's sterling efforts in responding to it.)

I think in the case where the OP explicitly acknowledges an answer as helpful it's more than reasonable to point out to them that accepting the answer is an option - even in respect of your own post - many new users miss that this feature exists (or forget it) and are often grateful for the chance to put their thanks to that poster in more concrete form.

Where there are several good answers that attract many upvotes, one might reasonably encourage the OP to consider choosing an answer. (I don't think it's reasonable to place the onus for encouraging others to understand the way stackexchange works only on the moderators. The responsibility falls to all of us.)

If you aren't only generating reminders on your own responses, but more generally encouraging people to 'follow the site culture', I think it's a very positive behavior.

  • 8
    $\begingroup$ An opinion about the second paragraph. I believe the best approach would be to upvote, award bounties and leaving comments emphasizing the good aspects of the answer, but not accept if it does not answer the question (at least if it does not answer any part of it; if it does 'ok'). It might mislead future readers and also discourage a new better answer to be posted. Agree in all other aspects. +1. $\endgroup$ – Andre Silva Sep 6 '13 at 11:56
  • 1
    $\begingroup$ @AndreSliva I understand that note of caution. By the time the tick was awarded it addressed enough that it wasn't really misleading to award it (it was okay as an answer, and in some ways a good answer to some of what I saw as side issues I raised, even though it failed to really answer the specific issue that caused the original post). I had an extended discussion in comments and the answerer responded multiple times with edits. It ended up as an answer of value to others that was worth more than +1, even though it left me unsatisfied in respect of my problem. $\endgroup$ – Glen_b -Reinstate Monica Sep 6 '13 at 22:27
  • $\begingroup$ -1 It may be annoying, but that is just tough. Also accepting an Answer that doesn't Answer the question is just wrong; who does that help? You can still award a bounty for effort, but not award the tick so people viewing the Q&A know the problem is not resolved. You never know who will stumble across the Q&A at a later date and provide then Answer you were looking for but may be put off contributing further by seeing the accept. $\endgroup$ – Reinstate Monica - G. Simpson Sep 17 '13 at 18:04
  • $\begingroup$ @GavinSimpson Looking at the help, it comes down to how you interpret the sentences "Choose one answer that you believe is the best solution to your problem" and "Wait until you receive an answer that answers your question well." - in particular, what constitutes the undefined 'well'. I feel that's a somewhat lower standard than the one for a bounty (the answer is 'exemplary'); the large discrepancy in reputation value seems to support that feeling. $\endgroup$ – Glen_b -Reinstate Monica Sep 17 '13 at 23:41
  • 1
    $\begingroup$ @GavinSimpson As for the annoyingness, what's annoying in particular is when the OP clearly indicates that they think an answer is good, and that it solves their problem ... but still doesn't choose it... and then simply disappears, since their problem is solved. I don't see the issue with reminding people of how SE works; it's not like we fear to remind people at other times. $\endgroup$ – Glen_b -Reinstate Monica Sep 17 '13 at 23:46
  • $\begingroup$ @Glen_b Re levels for bounty - there are several options available aren't there? You can award effort for example. The key word in the text you quote is "answers"; you shouldn't be ticking an Answer if it doesn't actually answer your question. Re annoyingness, yes, I agree in the circumstance you quote. That would be a good time for a timely reminder of how the site operates. Still, if the Stack Overflow experience is anything to go by I think it is better in general not to try to get users to accept beyond reminding them of the way the site works. $\endgroup$ – Reinstate Monica - G. Simpson Sep 18 '13 at 2:04
  • $\begingroup$ @GavinSimpson My discussion of the bounty reasons is based on this - which is what the help links to (under the subheading 'How do I start a bounty?'). There's only one 'good answer' reason, and the text given at the link above includes the word 'exemplary'. It's been a few weeks since I last did it, but my recollection is that the dialog-box when you offer a bounty to an existing answer is as suggested at the above link. $\endgroup$ – Glen_b -Reinstate Monica Sep 18 '13 at 2:18
  • $\begingroup$ @Glen_b Thanks, I think I was misremembering, but note the optional or additional custom text option. You could add it under the Exemplary option but add some custom text to explain the reward. $\endgroup$ – Reinstate Monica - G. Simpson Sep 18 '13 at 2:46
  • $\begingroup$ @GavinSimpson Yes I agree, that's certainly a possible action, but I think it's not actually any better justified than what I did. In the case in question, the custom text explaining the reasoning for me doing so would - were I to be truthful - pretty much have to be along the lines of "Even though I do think this has in one fairly clear sense been 'well-answered', and so by the official criteria could get the tick until a better one comes along, and even though I don't actually think it is 'exemplary', and so apparently excluded from this category of award, I'm doing it anyway." $\endgroup$ – Glen_b -Reinstate Monica Sep 18 '13 at 3:06

We sometimes encourage negligent posters to review how the site works and, in particular, to accept good answers if they can. This is a task that high-reputation community members often take on. When you politely address this general behavior it does the site some good.

A large number of the requests to accept an answer that I have seen come from people whose answers aren't that great in the first place: they are effectively rushing or even bullying the OP into a premature acceptance. People notice when you do this and if you do it too often you may acquire an "invisible" negative reputation. This amounts to a sense among the frequent voters in the community that you are in this only to drive up your reputation. It would only be natural for them to tend to cut you less slack and refrain from upvoting some of your marginal posts. (Some folks might enjoy doing a tongue-in-cheek Bayesian analysis of this phenomenon :-).)

If, after a suitable period of time (several days at least) you have posted an answer that, based on community votes, is clearly superior to the others, then you might post a comment that subtly reminds the OP of the value of marking one of the answers as accepted.

  • 9
    $\begingroup$ It will look like self-promotion if the user who answered the question will ask for an acceptance. It will look like an appropriate moderatorial activity when a heavyweight like whuber asks the OP the same. @Avraham, don't worry, such negligent behavior bugs others as much as yourself; everybody wants to see the answers accepted, so that a discussion comes to a reasonable closure. $\endgroup$ – StasK Sep 5 '13 at 15:26
  • $\begingroup$ Is it happening on this site? I haven't seen a large number of request to accept an answer, let alone questionable requests to accept one's own answer. $\endgroup$ – Gala Sep 8 '13 at 7:13
  • $\begingroup$ @Gaël Laurans Currently this happens rarely. There are intervals where we see more such requests, but it hasn't been a problem. $\endgroup$ – whuber Sep 8 '13 at 15:34

Yes; I also think it's reasonable to ask even if they don't comment—ask them to accept an answer or explain why it doesn't help them. But we do get a lot of users who appear once to ask a question & don't come back again once they get an answer.

[@whuber's answer includes caveats I hadn't considered (in fact I haven't noticed such bad behaviour on CV), but of course fully agree with.]

  • 3
    $\begingroup$ You're right: there's not much such bad behavior on CV. What little there is tends to come from people who frequent other SE sites or is associated with one or two high-rep users who no longer participate. (My theory is that when your principal motivation for participation is to accrue reputation, you won't last long here: the pursuit of reputation for its own sake is barren and ultimately comes to be perceived as such by almost everybody; you have to find rewards from the less-quantifiable effects of participation, such as seeing the community grow or watching the threads improve.) $\endgroup$ – whuber Sep 5 '13 at 15:57
  • 2
    $\begingroup$ I think it's a little presumptuous - without any input yet from the OP - to ask the OP to either accept the answer or explain why they don't want to. Who's the judge of whether the answer is so great that the OP must be crazy or just plain rude not to accept it? I think when the OP explicitly expresses gratitude for the answer, indicating it was exactly what they needed, the ambiguity is gone and it becomes purely a matter of etiquette (and not whether the answer "should" be accepted) for them to click the check mark. One exception may be for the much-loved "0% accept rate" folks... $\endgroup$ – Macro Sep 5 '13 at 16:22
  • $\begingroup$ @Macro: Well, given that enough time's passed, it's phrased as a polite request & not a demand, &c., it could be useful for the OP to say why the current answers are inadequate: perhaps he's found the answer elsewhere & will thus be encouraged to post it himself; perhaps he will explain something that was wrong with the answers & someone else will answer better; perhaps he's embarrassed to say he didn't understand. To my mind it's not an unfriendly thing after giving someone advice to follow up later with "So did that work all right for you?" - it shows a willingness to continue helping. $\endgroup$ – Scortchi - Reinstate Monica Sep 5 '13 at 16:56
  • $\begingroup$ @Macro I think that in the context of the question here, which indicates a situation where the OP has praised the answer, that it's quite possible to judge that even the OP thinks it's good. $\endgroup$ – Glen_b -Reinstate Monica Sep 6 '13 at 22:35
  • 1
    $\begingroup$ @Glen_b, yes I agree. When the OP praises the answer, I think reminding them of site etiquette re: accepting answers is perfectly appropriate (and I've done it before). I was mainly responding to Scortchi's statement "....even if they don't comment". $\endgroup$ – Macro Sep 7 '13 at 17:16
  • $\begingroup$ @Macro I see. Sorry. $\endgroup$ – Glen_b -Reinstate Monica Sep 7 '13 at 23:50

Quite apart from community norms and the optics of begging for reputation, accepting an answer also visually marks the question as “answered” (which means “no need to spend more time on it/check it” if you are looking for something to do and also “likely to provide a solution” if you are searching past questions to solve a problem you are having).

A lot of the discussion seems to focus on quality and reputation but in my mind the answer doesn't even have to be great or especially deserving to be accepted. If it provides a solution or somehow addresses the question (which might itself not be particularly deep or interesting), then the question ought to be marked as “answered” or the OP should explain what's missing.

Unlike upvotes, it's not so much about the reputation than it is about site management and good record keeping. I often ask people to consider accepting an answer (my own or someone else's) or to “upgrade” short comments to a full answer for this reason (it's not uncommon for a brief comment that seems unremarkable to its author to fully and completely address a question).

  • 2
    $\begingroup$ +1. As an aside opinion about the 1st part on 2nd sentence's 3rd paragraph (very specific :). If the OP explicitly says one answered his/her question I think it is very ok to ask for self-acceptance, but if he/she does not, then, I think this practice could backfire pushing the OP to accept answers when they do not want to (1 reason SE removed the former accepted rate displayed on users profiles). On the other hand, when users ask for acceptance on someone else's answer, IMO not only the answer was peer reviewed but it is also softer in the sense of putting some pressure on the OP... $\endgroup$ – Andre Silva Sep 9 '13 at 13:14
  • 1
    $\begingroup$ ...Totally agree about all other arguments used in favor of "good record keeping" and I think it was very positive to have brought this issue on the answer. $\endgroup$ – Andre Silva Sep 9 '13 at 13:15

Encouraging (pressuring - yes, it can be viewed as such) users to accept an Answer has been outlawed on StackOverflow for good reason. It is up to individual users whether they want to accept an answer or not. If they don't want to, that is up to them.

For example, the accept rate is now not displayed for a user in the Q&A to stop people berating users that have a low accept rate.

A timely reminder of how the SE sites work via a link to FAQ is OK for new users, but doing so for every unaccepted question would just clutter the site.

Other than that, vote on good Answers, downvote or comment on bad Answers, users visiting those Questions will be informed by the voting even if an accept is not forthcoming.

Finally, I don't buy into the idea that once an Answer is accepted the Q&A is complete. A better Answer may be forthcoming, or an alternative approach or solution provided. I often check back against interesting Q&As through the Favourite mechanism to see if there have been updates. CV, like the other SE sites, is an evolving beast, and that doesn't end with a click on the tick.


You must log in to answer this question.

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