3
$\begingroup$

This post is spurred by this question. The question has received 4 answers, all of them saying something different from the other and there is deep disagreement between all the answerers and the commenters. The comments were getting out of control and I have locked it to prevent any more back and forths. But let's take this opportunity to step back and look at the whole situation, so that we can avoid such instances in the future.

On the face of it, the question seems innocent and extremely basic. It's almost the kind where you'd want to jump on it to answer first and ride the upvote train. But wait... as a few users pointed out, the question really doesn't make sense, because a transform of the kind that the OP asked (reading the question literally), doesn't exist!

What do we do in this case? Do we read into the OP's mind and answer what we think that they should've asked? This is what all the answerers did. What if you're reading incorrectly? It would then be a waste of time for the person who wrote the answer, the few others who read it, others who commented that it is incorrect, etc. All of this because of an ambiguously worded post.

We must remember that while we're a welcoming site and will gladly answer simple questions or help people struggling to understand concepts, the question must also be written correctly and completely. If this is not done, please do not hesitate to close the question until it has been improved. That is, in fact, the very essence of the close/reopen system at Stack Exchange.

In such cases in the future, I suggest that we withhold answers until the actual question has been clarified either through comments or via chat and the question has been updated with the correct info. This is critical, because again, we can't expect people to read through pages of comments to see that OP changed their mind.

I'd be interested to hear your thoughts on this particular question and dealing with such incidents in the future. Please, let's try to keep this discussion about the nature of the question, rather than the technical details of it.

$\endgroup$
5
$\begingroup$

Does the question have one and only one likely interpretation?

If so, answer it, opening with "I think you meant..." so that readers understand the interpretation you are basing your answer on.

Does the question have a handful or fewer likely interpretations, the answers to which are either very short, or mostly similar to one another?

If the question can be answered completely and effectively without knowing the questioner's intention, go ahead and do so. This requires addressing each of the possibilities in a complete way, so it only works some of the time.

For example, "If you are on a Mac use $some_utility, $details. If you are on Linux use $other_utility, $details." is a reasonable approach if the user is looking for a software tools but didn't specify his/her OS.

Does the question require clarification to be answered completely and effectively?

In many cases, there are too many likely interpretations, or not enough information to determine what interpretations are likely. These questions just cannot be answered as-is. Post a comment to the question politely explaining what clarification is needed (you can't be too specific) and wait for the questioner to edit the question with more detail.

Is the question too vague to understand?

If you can't come up with a list of clarifications that fits in one comment (two if you are being very detailed in your request), or if the question has been around for a few days without any response to clarification requests, the question is too vague to be of use to the site as written.

Write an explanatory comment before voting to close (or upvote a request for clarification you agree with). If the poster seems like he/she might not know the ropes yet, including a "welcome" and a "your question may be re-opened if edited such that..." really helps communicate that closing a question isn't punitive, and that we aren't telling them to go away: we just need to improve the question in order to work with it.

$\endgroup$
4
$\begingroup$

I like HedgeMage's answer, but I don't think the question you point to fits with those options.

I think you hit the nail on the head with:

... the question must also be written correctly and completely. If this is not done, please do > not hesitate to close the question until it has been improved. That is, in fact, the very essence of the close/reopen system at Stack Exchange.

I didn't think about voting to close earlier because I wasn't aware that that was a valid use of the open / close mechanism (well, I didn't realize there was a re-open option until very recently).

$\endgroup$
  • $\begingroup$ Yes, closing merely prevents the addition of new answers. The post (and existing answers) can be edited, upvoted/downvoted as usual. So this gives a chance for the OP to fix the post, while at the same time, no one else has to attempt to answer an incorrect/incomplete question (only to find later that what they wrote is nullified by the new edit). Also, moderators can close/reopen with one click and so if the post has been edited to form, it can be opened sooner, if 5 community votes takes a while. $\endgroup$ – Lorem Ipsum Sep 26 '11 at 19:40
  • 1
    $\begingroup$ I think it would be useful, across all the SE sites, to update the "Why are some questions closed?" section of the FAQ with something to the effect of what yoda just said. As it stands, it still reads as if closing is much more severe/final (even though it technically says the same thing). $\endgroup$ – datageist Sep 27 '11 at 6:08

You must log in to answer this question.

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