5
\$\begingroup\$

We get a fair number of terribly phrased and formatted questions here. Here's a recent one from the Off-topic/unclear close pool. (Both Anko and I have tried to improve it, check the revisions):

https://gamedev.stackexchange.com/questions/70656/how-do-i-flip-a-sprite-along-an-angled-line

The original question was incomprehensible, but seemed like it could be salvaged to answerable by some edits. Unfortunately, the OP was so far afield that they hadn't even accurately expressed their intention. Once they received an answer, they explained further. As Krom said in comments,

That's always a problem with ill-spelled questions.

My question is: do we care what the OP was trying to say, or about forging good questions and answers from bad posts?

\$\endgroup\$

2 Answers 2

Reset to default
9
\$\begingroup\$

I think we do care, to an extent. The goal of SE sites is to provide an archive of good questions and answers foremost, but a secondary goal is to get individual users the answers they are after.

In cases like you've described, I would advocate for taking a firm hand in editing the question into something reasonable based on whatever context clues are available, provided that you can make a case for there being a single obvious question in there somewhere. Try, as best as possible, to keep the asker's intent in mind, but do prioritize the validity of the question over that intent if there's doubt.

Usually in these cases I'd leave a comment explaining that I edited the question but couldn't be sure I fully understood the asker, and request that they clarify my edit. This reinforces (especially to new users, who are usually the ones responsible for these kinds of questions) what happened to their question, that they can also edit, and various other details about the operation of the network.

However, if a question is really bad, to the point where there could be multiple equally-valid but unrelated questions in it, or you aren't confident what the intent is, please vote to close as unclear instead. Once a question has answers it's much harder to make major edits align without then doing somebody a disservice, so if that seems like it might happen it's better for the long-term health of the site to put the question on hold for a while rather than let ultimately unrelated answers clutter up the post.

\$\endgroup\$
2
  • \$\begingroup\$ +1 Reading this, I was enlightened. I've tried to follow Wikipedia's guidelines and be bold with edits: Accompanied with a comment explaining what I did and why, this initiates what Wikipedians call the Bold-Revert-Discuss-cycle, improving the post. However, this isn't enough: In contrast to WP, we distinguish Q from A. That is why we have holding and closure. We need good questions, but more importantly, we need them before answers. Questions that aren't there yet should be closed, then edited, then reopened. \$\endgroup\$
    – Anko
    Commented Feb 25, 2014 at 14:00
  • 2
    \$\begingroup\$ As a site, our primary value is in our answers (the questions are basically SEO for finding the answers). When there is a high-quality answer on a question -- whether or not it's the question the original poster was actually trying to ask -- I'd argue that our priority should always be to protect the answer, not the original poster's intent. \$\endgroup\$
    – Trevor Powell
    Commented Mar 7, 2014 at 4:09
5
\$\begingroup\$

I say, to hell with the OP's intent. Make good questions from bad.

The SE model is to make questions that are useful for future visitors, not the one person who posts one question and never returns. Good questions with good answers draw better views of the site, and serve as a good example for future posters. Bad questions that match a lone user's idea help no one.

The terms of service declare that a post becomes the property of the site, not the individual user. Even if that user does return, they can ask a new, better-phrased question to get to their precise problem. Or they can answer their own question, answered in the manner they intended, once they figure it out.

If the OP's ego is hurt by having their original post mutilated, then they ought to have asked a better question. Ego is also not guarded by the terms of service. Apologies to the casual users who are lost along the way. But those questions aren't what the site needs in order to prosper. The site needs good questions.

\$\endgroup\$
4
  • \$\begingroup\$ The moderation team on this subsite keep citing Stackexchange principles, but somehow, do not appear to embrace the most important idea behind it all. Perhaps it's a resource or community size issue. I'm sure, as a programmer you've found yourself Googling an obscure issue and are glad to find another, relevant, random query. Duplicates are one thing, as are questions that essentially beg for code, but questions such as these - stackoverflow.com/questions/1003241/… - Are the lifeblood of Stackexchange. And single shot questions ARE important. \$\endgroup\$
    – Charlie
    Commented Apr 9, 2014 at 22:53
  • \$\begingroup\$ I can't tell if you are agreeing with me or arguing against. What is the most important idea? Your example is a good general question; here's one from gamedev: gamedev.stackexchange.com/questions/16571/… I don't have any problem with this kind of question. I want there to be more of them. I think that removing crappy, individual-centered questions increases the likelihood of finding some random question with thorough answers. \$\endgroup\$
    – Seth Battin
    Commented Apr 9, 2014 at 23:47
  • \$\begingroup\$ I understand that the community is small when compared to Stackoverflow, and I can see why you'd not want the board overwhelmed with "how do I do this?", or "which one's the best?" questions. But this place takes it to an extreme. On Stackoverflow, you'll occasionally find questions with no answers. That's the approach you should take from time to time, I think... The idea is that someone will come across the query by happenstance and fill in that void, enriching your board. Either that, or the user/poster will get the hint, and in a large number of cases, self edit. \$\endgroup\$
    – Charlie
    Commented Apr 10, 2014 at 0:27
  • \$\begingroup\$ So I saw your question, and I interacted with it in the preferred way, which is to attempt to improve it. As moderator Josh suggested, you can always argue about close vote reasons on meta. Since you didn't ask specifically about your own question, i did it for you. Please participate. meta.gamedev.stackexchange.com/questions/1589/… As for the general tactic of what should remain open and what shouldn't, feel free to post about that here too. But be prepared to defend your case strongly. \$\endgroup\$
    – Seth Battin
    Commented Apr 10, 2014 at 0:58

You must log in to answer this question.

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