Is there ever a case where you would want to accept an answer, but not upvote it?
I can't think of a single situation where this would be the case.. So would it not make sense to have the "Accept" button upvote the answer too?
Since it is simple enough to cast an upvote and accept the answer, I don't a see a reason to constrain the user in this manner. I agree, I've never had a situation where I wasn't upvoting along with accepting. But, everyone has a different reason/rationale for upvoting, so I don't think the system should force that upvote.
I must have more than
40 answers (July 2009), make that 150 answers, (December 2010),
260 answers (June 2012), rather 408 answers (July 2013), or 603 answers (August 2014),
err 941 answers (August 2016), well... 1020 answers (July 2017), confirmed: 1169 answers (August 2018), indeed 1273 answers (June 2019) on non-wiki questions, accepted but not upvoted.
Mmmm, I see a trend here...
Those questions are often posted by "occasional" users with very few reputation points.
Most of them never accept an answer.
Some finally get the tick thingy, but completely miss the upvote ("I have accepted the answer, now, why would I be supposed to do something else?")
As I said in the meta question "What’s the single biggest barrier to entry on SO?", they simply do not know (and do not care).
Even the new (January 2013) about page helps only moderately (but it is certainly an improvement).
So would it not make sense to have the "Accept" button upvote the question too?
Even though I would benefit from such a feature, I am not in favour of that modification.
Accept is for accepting an answer.
Upvote is for upvoting an answer.
Keeping them separate seems right in term of UI design, following the "Don't make me Think" rules, as in "making things obvious vs. hidden information".
There's one corner case which hasn't been mentioned yet: you accept an answer, but you don't have any votes left for the day :)
Accepting w/out upvoting makes sense in one situation IMHO.
If you end up using an answer that you don't like, but works.
The check should not also upvote.
Following on from what VonC said. I see a lot of occasional users who might accept an answer, but don't have the 15 rep required for an upvote. Is this a special case where SO could apply the upvote when the answer is accepted?
There are badges available specifically for people who get a lot of their answers accepted with zero up-votes (Tenacious and Unsung Hero).
If the system were to change as suggested, to automatically give an up-vote to an answer you accept then those badges would have to change as well.
(this might not be a bad thing, as the current position does make winning those badges somewhat hit-and-miss, depending on the intelligence of the people you've tried to help)
Accepting an answer without upvoting it rarely makes sense (and never in the case of my own answers, of course). 8-)} But I agree with others that it shouldn't be forced or automated; having the system cast an upvote on your behalf seems wrong.
So how about this:
If you accept an answer and you're able to upvote it, clicking the checkmark could produce a pop-up prompt that asks whether you want to upvote it as well. Add some brief wording indicating that it's not mandatory, but probably a good idea.
Sometimes you get a number of answers, none of which quite hit the mark, but there is one that gets you 80% of the way to a solution. That's one reason for perhaps not awarding an upvote AND the correct answer.
The other reason is, as Jon points out, where you've run out of votes for the day.
Maybe some people think that +15 that you already get for an accepted answer is enough and they want to save their upvotes for other questions/answers? It's also possible that they're already out of votes for the day.
I personally don't think it makes any sense to accept an answer without upvoting it (upvote means "this was helpful", if it works and you're accepting it, it must have been helpful), but it's so easy to just click the other button I don't see any reason they should be combined into one atomic action.
Keep it separate. As Bill the Lizard pointed out, a lot of times users might think that "+15 is already enough". This is particularly true for the type of questions asked by new users (the ones most likely to not upvote when accepting an answer)... pointing out for the millionth time that someone's not using the correct operators shouldn't entitle me to 25 points... in fact, the easiest questions are already the ones generating the most upvotes as everyone feels qualified to agree.
I feel that one of the structural weaknesses of SO is that the way that rep is awarded rewards users who cherry-pick easy questions and rely on the laziness of voters who will only bother voting for answers that they can validate in 10 seconds or less... automatically awarding an additional +10 seems an unnecessary reward (at least this is my experience in the php/js/css tags where there are lots of inexperienced users)... there are a lot of technically correct but uninspiring answers, and providing even higher rewards for answering large numbers of simple questions provides exactly the opposite incentive that we should be aiming for... correct answers are rewarded well enough and can always be upvoted if the OP wants.
When asking questions, posting answers, voting (up/down), accepting, etc., we are participating of a community that helps us all.
What I personally value most is the legacy built by the community, which makes the system useful in the long term. An essential part of this is a rating system that reflects usefulness. I have benefited quite a lot from it.
I think that the focus on this usefulness for the SO community is often relegated, in favor of the reputation earned by users. Both go together, that is how SO works, but
When a decision is to be made, usefulness + legacy should prevail as a criterion.
Following up with this rationale, if accepting goes naturally with upvoting in most of the cases, something that proactively favors the upvoting in these cases would be most convenient.
Now, the due answer to the specific question: So would it not make sense to have the "Accept" button upvote the question too?, I would say YES, with an option for explicitly not upvoting. But any other action favoring this will also be interesting.
Note that, by setting the default for an answer being accepted to “Upvote” (as opposed to “Not upvote”):
It never makes sense to accept without an upvote, but the functionality should remain separate. It is a good way to judge people who actually 'get it' and deserve your attention.
I sometimes get two answers that are really close to each other in quality, and would like to accept both (but can't) — so what I'll do is:
I consider this to be a little more "fair" to the 2nd-place answerer, who gets 10 points from me. The accepted answerer is getting 15 points + the prestige of having the accepted answer.
When you've already upvoted an answer, but you come back later and realise that you also should have accepted it? (I've done this.)
I have several excellent reasons,
You can not upvote your own answer, leading to complications when a user answers their own question.
There are instances where the question might solve your answer, but does so in a low quality, or dirty (not best practice) manner that is not a good solution, although you do not mind using it yourself.
And finally, Nobody would be able to get the Neruto Hat then.
But seriously, as stated in other answers, it really is pretty easy to click the upvote button after accepting, so it wouldn't make much difference overall. While this was a great idea, I think it is not really worth the effort of the development team to implement.
Initially I thought this was a terrible idea, given how often I've seen answers that did, yes, solve the problem, but only barely and without much generality.
But the comments suggesting that the feature be limited only to initially upvoting (when possible), while still allowing an immediate undoing of this when desired, changed my mind.
Given that provision, there really is no technical reason it couldn't work well. If someone is out of votes for the day or doesn't have the rep to upvote? You just don't add the vote, simplicity itself. If someone doesn't really like the answer that much? Just undo the vote; the button will even change colors conveniently to alert you.
There are two remaining social considerations that I can see. The first is the expectation that would develop that an accept should also mean an upvote, potentially leading to drama in cases where that doesn't happen. Well, except that that expectation already strongly exists, so I don't see that changing much.
The second is a systemic shifting of careless accepters a little further toward upvotes than is presently the case, even for answers that are borderline. I'm not really sure this is a very serious problem, and arguably it's even part of the intention. (Especially for new users that just gained upvote privileges recently and aren't entirely aware of how things work.)
Given all this, it really seems like a reasonable feature to streamline things, especially if, the first time it triggers on a given account, a popup appears to explain what just happened.
Is there ever a case where you would want to accept an answer, but not upvote it?
Looking at the answers, comments, and votes on both, it seems most users here believe that when accepting an answer you are likely to upvote.
It may well be that in most scenarios, on some sites, the accepted answer is also worthy of an upvote from the question OP.
Some sites might have specific niches whereby an answer is either poor and worthy of nothing or a downvote, or good answers worthy of both being accepted and an upvote from the question OP.
But this is simply not "always" the case on all sites, in all tags.
Maybe it was more true 6 years ago when this was asked? That is perfectly possible and acceptable :)
So would it not make sense to have the "Accept" button upvote the answer too?
No, because the two functions serve entirely different purposes, and I'd like to retain control over being able to accept and upvote separately based on the given scenario.
Paraphrasing from another relevant answer of mine:
So there most certainly are scenarios where accepted and not upvoting is perfectly reasonable.
For example, in a "link only" answer, the information in the link's destination may well answer the question perfectly, so the OP can accept because "this answered their question".
But it being link only makes it a poor answer and so not worthy of an upvote.
(This is of course based on no other answers present, etc).
Or, perhaps more simply, the answer is just worthy of "accept" because it resolved my problem, but the answer author took no time or effort to explain when doing so could have been useful to me and others. Perhaps the grammar is bad, or the answer is generally lazy as if "here, I cannot be bothered, this will do you".
In which case I do not want to upvote, but accepting the answer is fair.
The point is, no matter how rare it may or may not be that we accept but do not also upvote, the fact there are scenarios where this happens means we should not automate it.
Here is an example of answer I intentionally accpected but did not upvote:
Why? The answer was good because it worked and answered the question...but I decided it did not deserve an upvote because it was rude and almost offensive (it has been edited since by a more clever person who removed the rude part of the answer, see the revision: https://serverfault.com/posts/744471/revisions).
I have previously not upvoted an answer that I accepted. I did that because the person who provided the answer did so in a comment, and SO won't allow you to accept a comment as a best answer, so I had to make a new answer myself that was basically just a copy of what he had said. It didn't feel right to upvote myself, especially when I wasn't the one who had come up with the answer.
The obvious solution here, I think, is to have the system automatically upvote when you accept, but also allow you to remove that upvote after acceptance.