On Stack Overflow it is considered acceptable to ask a question to which you know the answer, and then post the answer as an answer (not as part of the question). Similarly, it is acceptable to ask a question and then post a possible answer amongst others answers.

Here on UX, I have posted a design question. (Embarrassingly, I somehow posted it to meta here instead of the main site, but it's now been migrated.) In pondering the question I have come up with my own design ideas. When (and why) would it be appropriate to include these 'answers' in the question, versus posting them as answers for independent consideration?

I provided two possible answers to the question, and they have been deleted and edited to be part of the question. I'm less concerned about this particular question and answers, however, and more interested in general arguments regarding this behavior.

share
    
The only reason I personally would not want them as Answers is because it might deter others from answering, seeing that the question already has answers. That does not seem like the right reason to have them in the question, however. –  Phrogz Oct 8 '11 at 13:37
    
I'll preemptively argue that while this practice could result in more reputation for the person asking and answering (if the question and answers were upvoted), it is still a net benefit to the site, and ought not be discouraged based on this reason. While this is not (at all) why I thought it appropriate to post my own answers, it seems correct to me to award someone who has better 'designs' than someone else, and donates them to the site. –  Phrogz Oct 8 '11 at 13:41
add comment

2 Answers 2

I deleted your answers because someone had already incorporated them into the question. Other than that, I don't think there's anything generally wrong with answering your own questions

Having said that, I don't think that posting a bunch of answers to your own question to see which one "wins" is really appropriate. If your question is "which of these two solutions is better - or is there a better one", then post it as such, otherwise it's just rep games.

share
1  
I moved them into the question because they seemed like offered possibilities, not any sort of actual answer. If I were doing the same thing on SO I'd make my possible answers part of the question and wait for a "real" answer, if no one's solution is better than mine after a week or so I'll post an answer describing the way I solved the problem myself. –  Ben Brocka Oct 8 '11 at 17:01
    
@Vitaly If I was only soliciting "A or B" as the question and providing A and B, I totally agree. In the case where it's an open-ended design question with (presumably) a large number of possible answers, I suggest that adding two into the (potentially and hopefully large) pile of answers seems like a good seeding. Your thoughts? –  Phrogz Oct 8 '11 at 17:53
    
@BenBrocka I suppose that's the problem with subjective questions like mine: every possibility is an answer, and the only 'right' answer is what the majority votes up. –  Phrogz Oct 8 '11 at 17:54
    
@Phrogz Well, it's a large gray area, and everyone draws the line where they see fit. I personally wouldn't do this, I think, but it's really a matter of opinion. But where it comes to "seeding" answers - I don't think that this is a good practice for SE, and I don't think I've encountered it, or at least I haven't noticed. The question should seed good answers, not the OP's own answers. –  Vitaly Mijiritsky Oct 9 '11 at 20:50
add comment

If you want social proof to show you whether your suggested answers are better than those of your peers, you should be able to post them as answers. However, most often questions are not intended to do that; instead most people ask "I have a problem, and I'm thinking of solving it with A or B. How would you solve it?" In that case, it's likely better to keep your suggestions as part of the question.

share
add comment

You must log in to answer this question.

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