19
$\begingroup$

A user answered one of my questions with a very helpful, but incomplete response. With their hints I have completed the problem. Should I:

Accept their answer and leave it?

Accept their answer and post my own as well?

Post my own and mark it as the answer?

I'm not sure what the proper etiquette should be.

$\endgroup$
  • 1
    $\begingroup$ This looks relevant. It probably depends on the specifics but the last option you mention sounds like the best one if the other answer is incomplete. $\endgroup$ – Najib Idrissi May 23 '16 at 15:48
  • $\begingroup$ I'd accept the incomplete answer, but I might still post my own. $\endgroup$ – Michael Hardy May 27 '16 at 1:13
21
$\begingroup$

My own feeling is that a post that allows you to finish the problem yourself is the very best that can happen. I get disappointed when it becomes clear that the OP (the person asking) is not willing to follow my hints on how to finish.

As the saying goes: Give a man a fire and he's warm for one night. Set fire to him and he's warm for the rest of his life.

Glad you were able to do it yourself. Accept the answer and, if so minded, post your answer with the Community Wiki option.

$\endgroup$
  • 1
    $\begingroup$ Why post the answer as CW? $\endgroup$ – Najib Idrissi May 23 '16 at 17:10
  • 1
    $\begingroup$ @NajibIdrissi just general purpose if an answer is, say, commentary or detail on an answer by another. $\endgroup$ – Will Jagy May 23 '16 at 17:14
  • 12
    $\begingroup$ C'mon baby light my fire! Time to set the night on fire! $\endgroup$ – Asaf Karagila May 23 '16 at 18:19
  • 7
    $\begingroup$ I agree with the sentiment of this, but I would not insist on the CW option. But, this is the best that can happen to a hint answer! Often the OP would benefit from posting their own understanding of how to use the hint. That way they get feedback on their solution, and eventual gaps in their argument are exposed... $\endgroup$ – Jyrki Lahtonen May 23 '16 at 18:29
  • 2
    $\begingroup$ Pterry reference: mandatory upvote. $\endgroup$ – TRiG May 24 '16 at 20:30
  • 1
    $\begingroup$ "Give a man a fire and he's warm for one night. Set fire to him and he's warm for the rest of his life.", Perfect $\endgroup$ – A---B May 31 '16 at 19:38
  • $\begingroup$ People often focus so much on the "question - answer" aspect of Stack Exchange that they forget it's about solving practical problems. Sometimes complex problems are like an old stopwatch - just have to unstuck that one single gear and the whole piece starts ticking again!! I do have to agree with Jyrki though that CW is a poor choice. A better option would be to edit the answer with the completed solution and accept it. $\endgroup$ – corsiKa Jun 2 '16 at 17:49
  • $\begingroup$ I actually don't understand the relevance of the "fire" quote. It seems like an amusing non-sequitur. Shouldn't the relevant quote be more like "Give a man a fire and he's warm for one night. Teach him to build a fire and he's warm for the rest of his life." (As opposed to him being warm for the rest of his life because he soon perishes in flames.) $\endgroup$ – Cheerful Parsnip Jun 4 '16 at 18:34
  • $\begingroup$ @GrumpyParsnip You are correct about the relevant quote. Most people are so familiar with it that they will understand it even if the complete and correct quote is not provided. The comment about setting a man on fire is a darkly humorous twist. $\endgroup$ – Readin Jun 6 '16 at 5:47
  • $\begingroup$ When I'm looking for information on Stackexchange I'm really looking for an answer, not a hint. If the answer is difficult enough that hints can be given, then a good answer will have hints at the beginning to help guide the reader. If someone just wants hints they can stop after the first couple sentences and work the rest out on their own. My advice would be to edit the answer and accept it if you have enough rep to edit. Otherwise accept the answer and add the additional details in the comments. $\endgroup$ – Readin Jun 6 '16 at 5:55
1
$\begingroup$

I think it really depends on the reason the answer is incomplete. If the answer was incomplete because they wanted to beat everyone else to the punch, then it deserves an upvote but not an accept, in my opinion. Go ahead and post your own answer, but acknowledging the hints.

But since you say the incomplete answer was "helpful," it suggests a genuine desire to give a proper answer rather than a desire to be the fastest gun in the West. So in that situation I would upvote and accept, but also post my own answer, again, acknowledging the hints.

$\endgroup$
2
$\begingroup$

I would post and accept my own answer (and I have done so in the past). My original rationale was that the accepted answer shows up first in the list of answers, so having the most complete answer appear first will help future visitors get the whole picture more quickly. Since self-accepted answers no longer go first, accepting doesn't automatically put the complete answer first. Nevertheless, accepting your own answer allows votes to place your answer first if enough people consider it more useful than the incomplete one, and having an accepted answer marks the question as answered in various listings. So I'd still stick to this approach.

There are a number of exceptions, though:

  • If the incomplete answer outlines a general technique, and your own answer is an application of that technique to the specific question. Then other visitors with a similar but not identical question will likely consider the technique more useful than the actual application, so the incomplete answer may be more useful.
  • If the one providing the incomplite answer had a fairly low reputation (below 1k or so), I'd assume that the extra rep from an accepted answer may mean a lot to such a person, and award the checkmark to them.
  • If it is obvious that the person providing the incomplete answer put a lot of effort and time into it, and your own answer was easy to obtain building on that, then again I'd award the checkmark to the post where the core of the work was done, even if some small final step was missing.
$\endgroup$
  • $\begingroup$ Can't say I agree with this philosophy - if they gave you the piece of information you needed to finish the problem, with the rest being "exercise left to the reader" it's more than likely they could have put the rest but decided not to patronize you on it - they eloquently gave you the piece you obviously needed and let you fill in the blanks. They should get the credit, not you. $\endgroup$ – corsiKa Jun 2 '16 at 17:51
  • $\begingroup$ @corsiKa: To me it's not so much about credit as it is about making the Q&A as a whole most useful to future visitors. Most 1k+ rep users probably won't care too much about +15 there. And the formulation of the answer can give intellectual credit where it's due: “thanks to the very useful answer by …, I worked out that …”. $\endgroup$ – MvG Jun 2 '16 at 18:28
  • $\begingroup$ To me, it's about what's going to be most helpful to future visitors. The missing piece you needed is most likely going to the the missing piece future visitors need. That's the one that should get the most attention. $\endgroup$ – corsiKa Jun 2 '16 at 19:30
  • 1
    $\begingroup$ @corsiKa I disagree - the piece missing from the OP's proof is almost never the piece missing from a later reader's proof, unless that particular flaw is a common one. As Tolstoy said, "All happy families are alike; each unhappy family is unhappy in its own way." A complete proof is more likely to benefit future users than one calibrated to the particular mental state of the OP. $\endgroup$ – Mario Carneiro Jun 4 '16 at 16:36

You must log in to answer this question.

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