6

Here's the scenario:

I have recieved a few good answers for a [code-bound] question I asked. Since they all make different contributions, I would like to write my own answer combining approaches and snippets of code from two different answers.

Would it be acceptable to accept my own answer (nice redundance, by the way)? Or should I rather accept one of the other answers? Or perhaps tell one of the authors to add the contributions from the other answer?

| |
  • 6
    I think it makes only sense to accept your own answer if you want to tell the community something (like e.g. here). Otherwise you should probably give it to the one who made (in your opinion) the most important solution to solving the problem. You can add something to your question that explains how you are using the answer, or how you combine different pieces from different answers. – user121799 Mar 26 '18 at 23:53
  • 7
    I agree with @marmot. I think that compilation answers are almost always inappropriate. Everyone knows that you can only accept one answer, so pick the one that helped you the most, or the one that you think is the best written, or some other criterion that makes sense to you. – Alan Munn Mar 27 '18 at 2:00
  • 3
    I also tend to think that compilation answers are normally unnecessary. But I also believe that adding your "answer" to the question is even worse. So if you are absolutely convinced that the community benefits from seeing several answers combined, I'd go for the compilation answer. In general, however, I think it is fair to assume that people do not need to see several answers combined, they'll be able to do that themselves, especially if it is clear that different approaches are offered. – moewe Mar 27 '18 at 6:37
  • @moewe I agree, but what I intended to write was a synthesis of two answers, not just a compilation. A compilation would be something like "Here's one way to do this: [1] And here is another way to do it: [2]". Instead, what I wanted to do is, using the "core" approach from one answer, add in improvements on other aspects of the code from another answer. – Anakhand Mar 27 '18 at 8:16
  • 5
    My definition of 'compilation' is a bit wider and not only includes listing answers (with link or code), but also what you call 'synthesis' if the act of synthesising does not overcome significant obstacles in combining the two answers. I don't know if that is the case for what you have in mind. If anyone who reads the answers can combine the two in the way you have in mind without undue difficulty, I'd say a new answer is pointless. If you add value by putting two answers together in a non-trivial way, then go for it (that's what I meant with 'the community benefits' from your answer). – moewe Mar 27 '18 at 10:04
  • @moewe Ok, I understand now. Thanks for sharing your opinion! – Anakhand Mar 27 '18 at 16:17
  • 1
    You can also add your compilation answer as a postscript (edit) to the question. – John Kormylo Apr 1 '18 at 22:52
10

My opinion is (call it 'my' accepting -- etiquette, ;-))

  • If there is no answer yet and one has found a solution of one's own in the meantime, it is perfectly ok to accept the own answer. If someone else posts another (good) answer later on, it is very nice to unaccept the self-answer and accept the other one. (I've done so with one of my questions)

  • Instead of accepting a self - answer the relevant post can be transferred into a community wiki - answer as well, in order to allow for other answers to gain reputation.

  • If there are other answers but yet unsatisfying, it might be alright to accept the self-answer, but ask the other users to 'improve/change' their answers before, such that one of them can be accepted, but don't urge the users to incorporate content from other answers.

  • If there is or are (many) good answer(s) posting a compilation answer and accepting it is not really nice, since it is some kind of 'stealing' the good content from other ones. Don't do that.

| |
  • 1
    Yeah, I see what the general criteria are now (and why they are so). I've ended up accepting the answer I thought contained the best solution for the core problem. Other minor adjustments aren't strictly relevant to the "naked" question, but if visitors still want to look for them, they will find them in other answers. If I really wanted to point those out, though, would it be better to do so in an edit in the question's body itself or as a separate answer? – Anakhand Mar 27 '18 at 16:15
  • 1
    @Anakhand: As long as an edit of the question doesn't change its original meaning (and the question isn't closed already) such edits are alright, in my point of view – user31729 Mar 27 '18 at 21:05
  • 3
    @Anakhand I would say: don't edit your question to include an answer regardless. If you want to point them out, post another answer (maybe community wiki). Anything else is just confusing: the question space is for the question. Answers are not meant to go there. – cfr Mar 30 '18 at 3:24

You must log in to answer this question.

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