I asked https://stackoverflow.com/q/12434402, which was closed.

Why was this question closed? As far as I can see, the question seems to comply with all of the site's guidelines.

A moderator closed the question as "not constructive". The standard text for closing as "not constructive" says the question "will likely solicit debate, arguments, polling, or extended discussion", but I don't see how that applies to my question. I'm asking a practical, answerable, factual question. I don't see anything in my question called for debate, polling, arguments, or extended discussion. Can anyone help me understand what was wrong with the question and how to fix it or avoid this in the future?

Is it possible the moderator just clicked an arbitrary category, because none of the standard categories fully described the reasons for closing the question? If that is so, could I request an explanation of why my question was closed, and what guidelines it violated? If it is unsuitable as it stands, is there some change I could make that would make it more suitable for this site?

Edit 9/24: I got some good answers, and now I think I understand what was going on. I don't think I need any more explanations. Thanks, everyone, for your help and your time!

  • 4
    "if the standard category does not explain why the question was closed" If the standard category doesn't explain it, then the standard category isn't doing its job. Your problem is that you don't buy the explanation, not that the explanation isn't there. – Nicol Bolas Sep 24 '12 at 6:25
  • @NicolBolas, You may be right! I think my problem is that I can't even imagine how any of those phrases in the standard category would plausibly apply, so I assumed maybe it was closed for some other reason entirely. But maybe that was entirely the wrong assumption! Maybe this is a blind spot on my part. Any chance you might be willing spell it out for me? i.e., which part of the standard category explains it, and how/why does that part apply? ("solicits debate"? "arguments"? "polling"? "extended discussion"? something else?) Thanks for your comments! – D.W. Sep 24 '12 at 6:52
  • 3
    @D.W. Your question is also an exact duplicate. stackoverflow.com/questions/3701054/… Even if the close reason weren't valid (I believe it is), it's still been asked before. – George Stocker Sep 24 '12 at 10:56

Tool questions aren't fundamentally non-constructive

I disagree that questions asking for tools are inherently not-constructive. For example, while the poster here did ask for a tool, his real problem is not "I need a tool to..." but "...measure code coverage of a binary without source code".

I doubt people would say that "How can you measure code coverage of a binary without source code?" is inherently non-constructive. And yet it may very well be answerable with a "Here's a tool that does what you want." The others' arguments apply equally to that question, even though it is not asking for a tool. For that matter, they apply to a fair number of upvoted questions on SO.

Nothing wrong with questions that could possibly have multiple solutions

The argument seems to boil down to, "This type of question might have multiple valid solutions. Ergo all solutions are equally good, so non-constructive." That's the sorites fallacy. Posters are supposed to accept the answer most helpful to them, and it's perfectly fine for them to use their subjective judgement in choosing.

And besides, prima facie, for all we know there might be 0 or 1 tools that do what he wants. Looks to me that the task here is quite specialized, so I doubt there are many tools for it.

"Extended debate" reason is a red herring

The clause "this question will likely solicit debate, arguments, polling, or extended discussion." is supposed to prevent people from using Stack Exchange as a debate forum. Whereas a political debate may lead to continued interaction and go in circles, it doesn't look like this sort of question is likely to. It could be answered by, "Here's a tool, and this is how to use it to do what you asked." No need to debate any other answerers, and neither would I expect a debate to ensue.

Don't be trigger-happy

Finally, don't close just because it was formulated as "I need a tool to..." as opposed to "Does a tool exist to..." or "How can one..."; rather, edit it into shape. I have attempted to do so, editing out the explicit requests for tools that people seem to be hung up on.

That said, this question can be improved

In particular, it currently has little information about your setup. For example, what platform (Windows PE, Linux ELF, Java Native Interface) is the executable for? What language was it written in (tools for C vs. Haskell would be quite different)? Specifying these will help people answer.

(George Stocker found a duplicate, so I guess this is moot; it would be closed anyway. But the policy discussion is still relevant.)

| improve this answer | |
  • 1
    Thank you for your edits showing how you would improve the question! See your edits, and your comments above, set off a lightbulb in my head about why people may have had a negative reaction to the question (or the way it was phrased) -- as well as how to phrase this sort of question in the future so that it might stand a better chance of being suitable for this site. Much appreciated. – D.W. Sep 24 '12 at 7:59
  • @D.W.: Edited this answer with some specific suggestions. – Mechanical snail Sep 24 '12 at 8:08
  • 1
    "That's the sorites fallacy." And if this were a logical argument, rather than a discussion of what our rules are, that would matter. Our rules do not allow polling questions. A "find me a tool that does X" question is a polling question. We do not allow them. – Nicol Bolas Sep 24 '12 at 20:08
  • @NicolBolas: That's not the point. I contest that potentially having multiple answers automatically makes it a polling question. – Mechanical snail Sep 24 '12 at 20:16
  • 2
    @Mechanicalsnail: How is it not? In terms of answer quality and voting patterns, it is no different from asking, "What is the best tool that does X?" In both cases, everyone will post their tool of choice, and they'll upvote based on their preferences, not on anything remotely objective. – Nicol Bolas Sep 24 '12 at 20:17

You have a request for a list of tools you can try. You don't have a question that would elicit showing you how to fix a finite problem.

Given that your problem is that you need a tool and that you aren't using one right now, it's endless, wide ranging and that's not constructive.

Alternatively, flag your question to be re-opened and closed just as well under the "off topic" reason.

The constructive version of your question would be for you to already have started with a tool, any program really, and asked the gallery to leg you up over something you're trying to do with it.

| improve this answer | |
  • I apologize for being dense, but... I don't understand. I'm not looking for a list of tools (I'll edit the question to try to make that clearer). I'm looking for a single tool that meets the requirement. The problem is I don't know of any tool that meets the requirement. This is a finite problem -- finding a tool that provides coverage testing that does not require source code is a finite problem. Can you help me understand what is non-constructive or wide-ranging about that? Or can you offer any suggestions how I could revise the question to avoid this misunderstanding? – D.W. Sep 24 '12 at 3:11
  • 6
    It's the difference between "What car should I download?" and "What screw size is needed to re-affix the front passenger door to a Honda Civic?" – random Sep 24 '12 at 3:21
  • 1
    I'm sorry, I don't follow your analogy. (You say I should have "already have started with a tool". How am I supposed to do that, when I don't know of any coverage-testing tool that works without source code?) I'm honestly mystified... – D.W. Sep 24 '12 at 3:51
  • 1
    @D.W. If you has said, "I'm using tool XYZ, and doing ABC, and I'm getting error DEF. What do I need to change so that I can end up with the correct result of MNO" then it would have been an acceptable question. Asking for "what tool can I use to solve problem X" is more or less a shopping question – Servy Sep 24 '12 at 20:34

I am looking for a tool to ...

See that? That right there is where your question is "not constructive". It doesn't matter what comes after this.

Any question of that form, "Give me a tool that does X" is a list question. List questions are questions where everyone chimes in with their favorite "tool that does X", and then you arbitrarily decide that one of them is "right" and give them the acceptance check.

These kinds of questions are poll questions in disguise, because any tool that fits the "X" description can qualify as an answer. That's not acceptable here; no answer is objectively "better" than another (except those where the tool doesn't do "X").

| improve this answer | |
  • Do you have any advice on how to frame the question better? (Possibly relevant background: I'm not sure if such a coverage-testing tool even exists.) – D.W. Sep 24 '12 at 6:44
  • 1
    I disagree that questions asking for tools are inherently not-constructive. For example, while the poster here did ask for a tool, his real problem is not "I need a tool to..." but "...measure code coverage of a binary without source code". Don't close just because it was formulated as "I need a tool to..." as opposed to "Does a tool exist to..." or "How can one..."; rather, edit it into shape. – Mechanical snail Sep 24 '12 at 7:30
  • 1
    @Mechanicalsnail: "edit it into shape" No. It is not my job to take someone's garbage question and turn it into a good one. If he wants to fix his question and make it reasonable, that's great. But until then, it's junk, and it should be closed. You're basically admitting that it is a bad question. And we close bad questions. – Nicol Bolas Sep 25 '12 at 0:19

closed as not constructive by casperOne

As it currently stands, this question is not a good fit for our Q&A format. We expect answers to be supported by facts, references, or specific expertise, but this question will likely solicit debate, arguments, polling, or extended discussion.

Read out the reason for closing the post.

We expect answers to be supported by facts, references, or specific expertise

What the moderator is trying to convey that you should have some research work at your own, try to google this topic, gather some facts(might be some tools in your case), then ask about those facts. Without that, there is no point to start with.

this question will likely solicit debate, arguments, polling, or extended discussion.

This discussion has nothing to start with, so it will result in different views from different experts, which will ultimately result in arguments or long debates. So the discussion will never get to its end.

So it would be rather better if you first gather some facts on this topic, edit the question and flag it for reopening.

| improve this answer | |
  • I don't know this topic well, but it looks like the user did do some research, and explained why the solutions he found don't apply: "I do not have source code. I cannot recompile the program. (This rules out tools like gcov.)". – Mechanical snail Sep 24 '12 at 7:32

You must log in to answer this question.

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