13

I'm reading some questions that have no long-term value. A person posts about 100 lines of code that doesn't work. Not a technical issue. They just have a logic bug somewhere. However, they have carefully described what they did to try to solve it, and where they think the problem is.

I read FAQ about good/bad questions, and didn't see anything that would say "not a good question to ask here". After all:

  1. their question includes some code.
  2. they describe what is going wrong.
  3. they've described what they did to try to get it to work.

I'm wondering where to draw the line. Basically, someone is looking for free debugging help. But once they've been helped, the question and answer are not likely to be useful to anyone else.

Am I wrong in thinking this is out of place? If I am not wrong, is there something that could be added to the FAQ to clarify why such a question is not useful/appropriate here?

Improve this question
9
  • 5
    This sounds a lot like the old too localized reason for closing a question. You could add this to your title and/or text of your question.
    – Paul Hiemstra
    Jul 10, 2013 at 12:10
  • 3
    Yes, many of these questions should be discouraged. Downvote and closevote them.
    – Pekka
    Jul 10, 2013 at 12:11
  • 1
    As long as they made a good effort to describe what's wrong, and to understand the problem, I don't mind these kinds of questions. In all likelyhood someone else will come across a similar error message and the question/answer will be useful to them even if it is fairly localized.
    – Mansfield
    Jul 10, 2013 at 12:11
  • 2
    If it was an error message, I might consider it okay. This is just screwed up code, that produced the wrong result.
    – ToolmakerSteve
    Jul 10, 2013 at 12:12
  • @ToolmakerSteve Do you have an example question?
    – Mansfield
    Jul 10, 2013 at 12:15
  • 2
    IMHO, the FAQ really should address this topic head-on. Don't need these "beginner programmer -- debug my code for me" questions, even when well asked.
    – ToolmakerSteve
    Jul 10, 2013 at 12:16
  • @Mansfield stackoverflow.com/questions/17568297/…
    – ToolmakerSteve
    Jul 10, 2013 at 12:17
  • @ToolmakerSteve Yeah, that one isn't great. But I think there are some situations where similar questions would be acceptable. I'll see if I can find an example.
    – Mansfield
    Jul 10, 2013 at 12:22
  • @ToolmakerSteve In your opinion where should be it placed if not on Stack Overflow. It all comes to old discussions, like Why are we rude to new comers or Is SO reached saturation. I agree that one you linked wasn't good but it deserves a stand unless too localised
    – CRUSADER
    Jul 10, 2013 at 12:30

1 Answer 1

Reset to default
6

These questions (on the whole) are welcome, so long as they meet several criteria (and these are criteria that apply to every question, but bear repeating):

  1. They must have a good title that explains the issue (hint: "What is wrong with my code?" is not a good title)
  2. They should contain an SSCCE (Part of the SSCCE is "Short". A code dump that scrolls is not short)
  3. They should include the error message. This is what future users will search for when they have a problem.
  4. The user should explain what they've tried, what hasn't worked, and what they expect to happen.

If these criteria are followed, then there's a good likelihood that future users will benefit both from the question and its answers.

Keep in mind, this is not for every syntax error or typo question out there. There are still instances where no matter how the user phrases the question, it won't be helpful to others. I address this below.

Users search on error messages, and any answers that explain the error message in depth and how to fix it are good answers. We can't have good answers if we don't allow people to ask questions because their error message indicates a syntax or logic error.

There are several ways these types of questions can be closed, depending on what's wrong with them:

  1. No good title or just a long code dump with a "What's wrong with my code?": -> off topic

    Questions concerning problems with code you've written must describe the specific problem and include valid code to reproduce it. See SSCCE.org for guidance.

  2. Short code dump, but the user forgot how to spell their own name?. I argue they didn't have a minimal understanding of the problem: -> off topic

    Questions asking for code must demonstrate a minimal understanding of the problem being solved. Tell us what you've tried to do, why it didn't work, and how it should work. See also: Stack Overflow question checklist

  3. Can't figure out what they're even asking for? ->unclear what you're asking:

    Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking.

The reason 'Too Localized' was taken away was because it was being used on questions it should not have been used on way too often. Questions with syntax errors are a part of programming. It is inappropriate for us to close questions that meet the criteria I listed for a good question, even if it's because the OP made a stupid mistake.

The good news is that most 'typo' questions that meet my criteria have been asked before, if they have a good answer, feel free to mark them as duplicates. Of course, the whole duplicate question thing is a whole 'nother issue.

Keep in mind, we want Stack Overflow to be a repository of useful programming knowledge. That means that users shouldn't have to go to forums to find out what error messages mean.

Improve this answer
12
  • What close reason would you suggest to use on this question? (The sub reason, not the general "Off-topic" reason. None of the reasons on the next screen seem to be guite right. stackoverflow.com/q/17568297/1324019
    – Mansfield
    Jul 10, 2013 at 13:08
  • @Mansfield That's not a 'typo' question, that's a logical error. In this case, I would have closed it because the user didn't narrow down the code dump to the least amount needed to reproduce the issue (if they had, they likely would have solved the issue themselves).
    – George Stocker
    Jul 10, 2013 at 13:11
  • I guess that falls under the "valid code to reproduce the problem" close reason. But even then, that doesn't strike me as the real reason why that question isn't good.
    – Mansfield
    Jul 10, 2013 at 13:12
  • I didn't say anything about "error messages" in my original example. Nor in the link I posted when @Mansfield asked for an example. I'm talking about beginner programmers, who don't know how to debug their own code. Its just buggy. It has some logic error. They need help debugging. Has zero relevance to anyone else on the planet. the FAQ says NOTHING to discourage such a question. Why not? And what close reason would make sense to this newcomer? Not make sense to oldtimers, but to a newcomer?
    – ToolmakerSteve
    Jul 10, 2013 at 13:23
  • @ToolmakerSteve I answer this in my post implicitly: If a user posts a question, and that question does not meet any of the guidelines for closure, it should be open. In some cases (specifically in that question you mention; it could either be closed because the user did not have a minimal understanding of the problem, or because they did not post an SSCCE (emphasis on the 'Short' part of SSCCEE).
    – George Stocker
    Jul 10, 2013 at 13:26
  • "we want Stack Overflow to be a repository of useful programming knowledge." I am saying that here is a question that clearly does not contribute to useful programming knowledge. I've read your answer, I do NOT see how it relates to what I am asking.
    – ToolmakerSteve
    Jul 10, 2013 at 13:28
  • To Clarify: The question is reasonably well written (re SSCCE). But the actual code, and the problem with it, seems unlikely to have any relevance beyond the original posting. So, why is it here? Its just clutter. I don't think SO is here to help everyone who has a debugging problem. Only if they've isolated it into a form, the solution of which, might be useful to someone else. Do you disagree with this?
    – ToolmakerSteve
    Jul 10, 2013 at 13:33
  • @ToolmakerSteve I disagree that the code is reduced to the least amount needed to reproduce the issue (the 'short' part of SSCCE). If the user took the time to do that, they'd likely find the error themselves.
    – George Stocker
    Jul 10, 2013 at 13:39
  • As Mansfield said "I guess that falls under the "valid code to reproduce the problem" close reason. But even then, that doesn't strike me as the real reason why that question isn't good." -- Exactly my point. We DON'T have an APPROPRIATE close reason. So, as it stands, it shouldn't be closed. But this makes no sense to me. That's why I am going on about it. I either need to understand WHY it shouldn't be closed, or WHY I am wrong about the lack of an appropriate close reason. OR, I might be on to something (that has come up before..)
    – ToolmakerSteve
    Jul 10, 2013 at 13:40
  • @George. The poster seems to have made an effort to not post an excessive amount of code. They are doing SSCCE, within the limits of their (obviously very limited) abilities. I just plain think they shouldn't be asking the question at all. They should keep trying to debug their code, or go to some other type of site for help. But I think slapping one of the existing close reasons on their post would not be educational to them, or anyone else. What they need, is to learn basic debugging skills. I've said something along those lines in a comment in their post. I'll leave it at that.
    – ToolmakerSteve
    Jul 10, 2013 at 13:43
  • @ToolmakerSteve Your best bet then is to ask for a feature request to add a new close reason that adequately describes what you're trying to prevent, with examples (more than 1) of questions that meet this criteria and how they aren't useful to future visitors. You're asking for discussion here, and I'm very plainly telling you that under our current system, these questions are welcome.
    – George Stocker
    Jul 10, 2013 at 13:45
  • @ToolmakerSteve That having been said, my personal opinion is that we should find a way to improve the close reasons to shut down questions where the user needs to debug and there's no way the question is useful to future visitors. If you had a well worded feature request with examples, I'll be glad to vote for it.
    – George Stocker
    Jul 10, 2013 at 13:48

You must log in to answer this question.

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