8
$\begingroup$

I asked a question on MO and it now has three correct answers, all more or less using the same strategy and similar to each other. If I accept all of them, do they all get the reputation they deserve?

$\endgroup$
1
  • 8
    $\begingroup$ MO allows only one answer per question to be "accepted". Thus you need to decide. $\endgroup$
    – Stefan Kohl Mod
    Commented Sep 17, 2013 at 20:07

3 Answers 3

14
$\begingroup$

As others have stated, you cannot accept more than one answer.

Don't worry about reputation too much, I think almost no one cares if they get 15 points more or 15 points less than others.

Accept the answer you like most, and if they all look the same to you go by some arbitrary rule to break ties like the chronologically first answer, the answer with highest vote, etc.

$\endgroup$
2
  • 6
    $\begingroup$ +1 for "Don't worry about reputation too much, I think almost no one cares if they get 15 points more or 15 points less than others." I'm not sure whether that's true in practice, but I hope it is. $\endgroup$ Commented Sep 18, 2013 at 7:27
  • 3
    $\begingroup$ "I think almost no one cares if they get 15 points more or 15 points less than others." You'd be surprised how much people care about internet points. $\endgroup$ Commented Sep 20, 2013 at 5:54
10
$\begingroup$

In absence of good reasons to prefer one answer, sometimes people choose the first given one, and leaves some nice comment to the others.

(Look e.g. here: I was honored to have the same example as George Lowther, even with the same notation, one minute later).

$\endgroup$
4
  • 7
    $\begingroup$ Do you have evidence to support the assertion that "usually people choose the first given one"? I could just as easily assert that people usually choose the one with the highest reputation, so as to curry favor with the bigwigs; or that people usually choose the one with the lowest reputation, to reward the one most needing the points; and so on. One example is not enough to support "usually". $\endgroup$ Commented Sep 18, 2013 at 0:09
  • 1
    $\begingroup$ I have observed it in almost all cases when two ore more answer are very similar, cases which are not that frequent however. But yes, I had the feeling that "usually" was too strong claim. MOre reasonable is "sometimes". $\endgroup$ Commented Sep 18, 2013 at 6:03
  • 3
    $\begingroup$ On purely anecdotal grounds, my experience is that indeed the first one given often wins the big points prize. (This effect seems much more pronounced over at MSE, which I've been spending time at recently, and which seems insane by comparison -- people constantly racing to be the first respondent, and with never-ceasing questions continually descending like snowflakes from the sky. It's really a mess over there!) $\endgroup$
    – Todd Trimble Mod
    Commented Sep 18, 2013 at 16:54
  • $\begingroup$ My experience is that often if one of similar answers posted at a similar time gets an early one-point lead because of some random fluctuation, it is likely that people will continue to vote it up and it will get eventually accepted, presumably because it appears on the top of the page and people read it first. This is irrespective of which answer was actually given first. $\endgroup$ Commented Sep 19, 2013 at 15:31
-8
$\begingroup$

Since you can't accept more than one answer, a common method to approximate fairness is to accept one, and vote up the other correct answers. One person will get 5 more points than the others.

$\endgroup$
17
  • 3
    $\begingroup$ That's weird, why are people downvoting this? $\endgroup$ Commented Sep 17, 2013 at 23:09
  • 8
    $\begingroup$ @ScottMorrison Maybe to increase the difference to 10? $\endgroup$ Commented Sep 17, 2013 at 23:55
  • 11
    $\begingroup$ My guess is that people don't like the downvoting suggestion. But I'm not sure. $\endgroup$
    – Todd Trimble Mod
    Commented Sep 18, 2013 at 0:01
  • 1
    $\begingroup$ @AndresCaicedo, voting on meta has no reputation effect. $\endgroup$ Commented Sep 18, 2013 at 0:03
  • 2
    $\begingroup$ @ScottMorrison Which is both true, and a non sequitur. $\endgroup$ Commented Sep 18, 2013 at 0:09
  • $\begingroup$ @AndresCaicedo, I'm confused: wasn't your suggestion that people might be downvoting to increase the difference in the points that the two answerers might receive to 10, and I replied that they couldn't change that difference, because it would always be zero? $\endgroup$ Commented Sep 18, 2013 at 0:11
  • 19
    $\begingroup$ @ScottMorrison Not seriously, the difference in this case is in score, not in reputation. Seriously, the proposal of downvoting a correct answer is absurd. $\endgroup$ Commented Sep 18, 2013 at 0:13
  • 6
    $\begingroup$ Can one of the mods please delete this suggestion (instead of merely rephrasing it as absurd) so that it does not come across as officially-approved? $\endgroup$ Commented Sep 18, 2013 at 3:06
  • 4
    $\begingroup$ It is not just that the suggestion of downvoting seems mean. To me, anyway, the problem here is that reputation is meant to be a broadly aggregated guide to community feeling, and some fun, and both of those purposes are damaged when it becomes a target for manipulation. $\endgroup$ Commented Sep 18, 2013 at 12:04
  • $\begingroup$ @ColinMcLarty I do not understand your comment. The intent of the suggestion seems to be to distribute reputation points in as fait a manner as possible. (Equal answer should receive equal credit, yet one 'must' be accepted, hence this suggestion.) So, on a reputation scale this suggestion in my mind tries to precisely minimize the 'manipulation' intrinsic to the fact that sometimes there is no one best answer and still one should be marked as such and get some then perhaps unfair bonus over the others. $\endgroup$
    – user9072
    Commented Sep 18, 2013 at 21:45
  • 1
    $\begingroup$ But, I would say an issue with this suggestion is that answers score should reflect community approval of an answer (while accept refelcts OP's opinion), and the score gets somehow biased in this way. Since this biasing the score is relatively stronger, I think it is more of a problem and thus do not consider the suggestion as very good but rather think voting and accepting should be done rather independently. $\endgroup$
    – user9072
    Commented Sep 18, 2013 at 21:45
  • $\begingroup$ There also is an equality/efficiency tradeoff. We might think it is good that good answers are rewarded and that people should get equal reputation for equal contributions. If we think the former aspect is more important, we might reject this suggestion (even after the edit). $\endgroup$ Commented Sep 19, 2013 at 9:30
  • 1
    $\begingroup$ @AdamP.Goucher clicking on "edited [some time]" in the middle at the end of the answer you can see earlier versions of the answer, which did mention the possibility of downvoting the accepted answer to blance point-distribution. $\endgroup$
    – user9072
    Commented Sep 19, 2013 at 11:43
  • 1
    $\begingroup$ @quid Yes the intent of the suggestion is to enable posters to distribute reputation in what the poster deems a fair manner. But I think the intent of the system is to aggregate the community feeling in some uniform way, giving the poster's view more weight than others but not as much weight as the suggested procedure would. $\endgroup$ Commented Sep 19, 2013 at 13:58
  • 1
    $\begingroup$ @ColinMcLarty but in my mind in the suggested procedure the porster's view has rather less weight than it could/does have in a more standard situation. Suppose the suggestion would have been: "just upvote all the (good/equivalent) answers and accept none" [which is a way some take I think in some situations, especially in CW; and would not unreasonale in my opinion as there is no urgent need for accepting an answer, though it typically should be done] Would this also be giving more weight to posters view than the intent of the system. (Current sugg is approx to this, while still accepting) $\endgroup$
    – user9072
    Commented Sep 19, 2013 at 20:07

You must log in to answer this question.

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