Revisiting closure, part 2: What are good motivations for closure?

Question

What is this and why does it need to be discussed?

An illustrative example

Consider the following discussion on the closure of a question:

• User A: I am voting to close this question as a duplicate of question X.
• User B: Why? Question X is obviously different.
• User A: The answers to question X answer this question.
• User B: This question is asking for […], but question X is asking for […].
• User A: Look at answer Y. It directly provides an answer to this question.
• User B: Yes, but this is not what is asked here.
• User A: Talking to you is like talking to a brick wall.
• User B: Your father smelled of elderberries.

This discussion leads nowhere and escalates because User A and B are arguing on the wrong level. They agree about each other’s assessments (“the answers to question X already answer it”, “question X is different”), but argue about them. However, they fail to see that they disagree on a more general level, namely what suffices for closure as a duplicate:

• User A focusses on similarity of the answers.
• User B focusses on similarity of the questions.

Our general problem

Going by previous discussions, comments left by close voters, and in custom close reasons, I think that we have a problem similar to that in the example: We discuss on a rather specific level; we argue from different motivations on a more general level; but we never address those differences. Even worse, I have the feeling that some people think that some motivations are carved in stone, even though they do not agree with them.

Before discussing about closure in detailed cases, I therefore wish to establish or confirm a common ground, i.e., a set of good motivations for closure.

Actual question

What are good, sufficient motivations for closure on this site?

Please mind the keyword sufficient: Naturally many of the above reasons tend to coincide, but that does not mean that they are good motivations for closure on their own.

• Each answer should only contain one motivation.
• Answers should be roughly one level more basic than a pre-defined close reason. If possible, elaborate why it is acceptable to close a question with this motivation and provide examples.
• Each answers should elaborate whether it is acceptable to answer questions fitting the motivation (i.e., that one votes to close with this motivation).
• As duplicate closure is considerably different from other reasons for closure and there seems to be no major disagreement about it, I would exclude motivations specific to it from this discussion.
• Indicate your agreement or disagreement by voting.
• To keep things simple, please refrain from playing devil’s advocate and posting motivations that you disagree with.

Potential answers

To provide some example answers, I will post some motivations that I consider good.

Furthermore, here are some motivations that I observed being given for closure (not that I agree with them):

• The question is better suited for/belongs on another site.

• The question has a low quality.

• The question cannot be answered by us.

• The question is not about design per se.

• The question can be answered by a simple Internet search.

• Answers to the question only help the asker.

Answer 1

‘Gimme teh codez’

This is a phrase which has been used for many years on StackOverflow especially to refer to questions which show no evidence that the asker has done anything at all to try to arrive at an answer, or at the very least a proper understanding of the problem, before asking. It is characterised by askers simply wanting full, working code that they can copy-paste into their own project without having to worry about understanding how the code works or what it really does.

It has several guises on different SE sites, but it boils down to the equivalent of asking for a fish instead of asking how to fish.

Here on GD, it mostly comes in the guise of people who find an image somewhere on the Internet and seemingly come straight here to ask, “How do I make this?”—or for that matter, “What is this font?”. The underlying question itself isn’t necessarily bad, and there can be a lot to learn from it, but the way it is asked reveals an attitude towards the community that we shouldn’t want to encourage. The most egregious questions of this type are ones that simply just ask people to do free work for them, like this one; these are quite consistently quickly closed, but there are less blatant ones that are sometimes closed, sometimes not.

This question is a recent example, posted yesterday. The entirety of the question reads:

https://dribbble.com/shots/3665378-Renault

Anyone knows how to achieve this in PSD?

As the answer by user287001 shows, there is a great deal of very useful detail to take from this question, so it definitely belongs here. But the basis of the question (shown also in the comment to the answer, asking for the PSD file) is a request for a finished, cut-and-paste job, rather than a request for an explanation of a way to solve the problem; a fish rather than a fishing rod.

In the same way that we require people to first try to identify fonts themselves by using automated font identifiers, it’s only reasonable to also expect people who ask for advice on how to achieve an effect, style, or design to show what their own understanding of the problem is and how far towards the goal that takes them.

If nothing of the sort is evident from the question, it should be perfectly acceptable to close it—keeping in mind that closing a question is not actually closing it, but putting it on hold pending improvements. (As Wrzlprmft mentions in the comment below, there actually used to be a tailored close-reason specifically for requests for free work.)

Answer 2

Our community cannot judge answers to the question

A central aspect about the Stack Exchange system is that answers are evaluated by users via votes and thus good answers raise to the top. This allows the askers and others to see which answers are judged to be the best by experts (i.e., us) and allows them to read the best answer first. If we cannot evaluate answers, this mechanism is broken.

In contrast to this similar answer, this is about questions which can be evaluated in general, but we are not not the experts but as good as random people from the Internet. Of course, for almost every good question, you will find some user of this site who can adequately judge it due to their hobbies or day job, but not because they care about graphic design.

Examples are out-of-scope questions, including a considerable portion of tech-support questions. For a specific example consider the question:

Does the uncertainty principle go against chaos theory?

Some of us may tell a good answer to that question from a bad one¹, but that has nothing to with us being interested in graphic design. We, as a community, are as good at evaluating this question as random folks from the Internet.

It is acceptable to answer such questions as the answers may help the asker and future visitors and are preserved in case of a possible migration. However, we should be careful not to encourage people to ask such questions here.

^{¹ in fact, I answered this question on Physics SE}

Answer 3

The question does not fit the general Stack Exchange format

These are questions that would not be a good fit for any Stack Exchange site¹.

Questions not fitting Stack Exchange’s question-and-answer format at least cause frustration because users cannot act as they want to or as is appropriate. Moreover, our mechanisms for improving the quality of content, getting rid of bad-quality content, making posts useful to future visitors, and moderation are impeded.

Typical examples of such questions are most opinion-based and too-broad questions as well as bad critique and style-identification questions.

It is not acceptable to answer such questions, as answers only increase the problems caused by such questions.

---

^{¹ with the possible exception of “special” sites such as Software Recommendations or Code Review}

Answer 4

There is no good criterion to tell a good answer from a bad one

A central aspect about the Stack Exchange system is that answers are evaluated by users via votes and thus good answers raise to the top. This allows the askers and others to see which answers are judged to be the best by experts (i.e., us) and allows them to read the best answer first. If we cannot evaluate answers, this mechanism is broken. Note that this is not about whether existing or expected answers to a question are good or bad. This is about questions where we cannot even conceive how to tell a good answer from a bad one.

Examples for such questions are unclear questions or primarily opinion-based questions. For a specific example, consider the question:

How did you start with Graphic Design?

As long as I answer this question truthfully, my answer is correct. Moreover, every (truthful) answer to this question is equally valid.

It is not acceptable to answer such a question as answers to such questions are a considerable source of dispute. Also, any answer may turn into a bad one if the question is salvaged.