What should we do with questions that don't show evidence of prior research?

Question

Recently a translation question was asked ("need translation of latin quote, pacis puella") that didn't show any research effort. This lead to a discussion in chat, where the question was raised – which close reason best applies to such questions?

I suggested "unclear what you're asking," but the text associated with that close reason is:

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.

That's sort of accurate – we did want additional details. But I believe the bigger reason for closing it was that it didn't show any effort; that is, no evidence of previous research was provided.

Up to now, we have consistently encouraged users to show their work; this was the point of the meta post How can I ask a translation or homework question? from very early in the site's history. But we have not extended that encouragement into a guideline or policy. That is, we haven't said, "If you don't show your work, your question will be closed."

My question here is, do we want to do that? Do we want to define some basic criteria for what kind of research must be done and communicated in new questions? Let me suggest that we focus on these points:

• Do we want to formalize a guideline in which we close questions that don't show research?
  • If so, which types of questions would fall under the new guideline? Translation only? How about basic grammar questions? How about history of Latin questions?
  • If so, broadly speaking, what kinds of resources would we expect users to check for each type of question?

If we go forward with this, we'll need to address other matters, like exact wording of the close reason and a resource list. But there's no sense in trying to cross that bridge before we get to it.

And finally, at the risk of making this question unbearably long, let me suggest the pros and cons that come to mind:

• Pro
  • A guideline makes it clear to the community what our standards are for questions that don't show research, which allows us to be more consistent in closing/reopening questions
  • Establishing a guideline like this allows us the define a custom close reason, which would be more helpful to new users, as it would explain more exactly what they need to do to have their question reopened.
• Con
  • If it's not done carefully, it can lead to many beginner questions being closed. Other Stack Exchange sites have guidelines like these, and they close many such questions.

So, what do you think? Let's discuss and see if we can come to agreement on what is in the best interest of this site!

Answer 1

I understand the desire to formalise such concerns. But I still think it is a solution in search of a problem.

It is my personal opinion that bureaucracy should always be minimalised: the fewer rules, the better. I would only invent new rules if:

• There is a serious problem;
• The problem can only reasonably be solved by creating this rule.

Is this really a serious problem? I have only seen a few poor questions of the same type, and in fact I think "unclear" is appropriate enough as a reason to close it.

As to "you don't show that you have done your research" as a closing-reason: "not enough research" is only a supporting condition to close certain kinds of questions, not a sufficient condition to close any question. The actual, complete closing reason for the question above should therefore be, "because it is unclear what you're asking, as you don't provide enough context, your question has been closed; if you had shown that you had done research, it might have remained open, but that is not the case". That seems a tad long. But the most important component of the reason is "unclear because too little context", not "no research".

So "your question has been closed because you don't show any research" suggests that no research is a sufficient condition, unless you add several clauses to the closing-reason, even if this is not what was intended by it.

On English Usage & Language, a similar reason was invented, with a very narrow scope: only questions that could be easily answered by consulting a standard dictionary, or a beginner's grammar book, were to be closed. But the use of this reason was soon expanded by those who wanted to close questions that they just deemed of low quality, and now it is used seemingly at random, or so it seems to me.

The more rules you have—especially if they are complex, broad, or subjective—, the more arbitrariness there will be.

Answer 2

I apologize that this answer is so long.

I think we should add a new closing reason for question with insufficient research. The main goal is to help users improve their questions. There would be no real changes to our rules.

Requiring research is already a policy

We already require research effort for questions. Some of it is based on our scope description, and some is built into the SE system. Perhaps the question is to what extent we ought to enforce this policy.

Our help page on on-topic questions tells that you should not ask translation requests or etymological queries that show no serious research effort using readily available online sources. (This help page is editable. There is a separate meta post about editing it.) We already have a policy that research effort is required.

Our help page on reopening questions tells that if your question has been closed, you should be sure that you've read the close notice and any comments on the question so you can address any concerns raised there. We cannot edit that help page, but we can make that bit of the page more helpful by making the close notice more helpful.

There is also a help page for quality standards errors, which tells that a question should include any background research you've tried but wasn't enough to solve your problem.

The how to ask help page also makes a clear point that research is needed.

Our tour also tells that you should not ask questions that show no research effort using readily available online sources. The tour is editable. (We might want to update the link in the tour if we have a better meta source for research requirements.)

Do we need to be more explicit about research?

I do not think so. We can provide more help, but the current rules suffice. I think it is already sufficiently clear that some research is required. And if we add the new closing reason, we can communicate the requirement to those who missed that line in our tour.

I agree with Cerberus that we should have no unnecessary rules or bureaucracy on the site. Requiring research is already a rule, and I would not make changes to the rule. I have nothing against tweaking the rule a bit or discussing how it should be enforced, but I would not make major changes to the rule itself. In my view the new closing reason would not add a rule, it would only add a means of communication.

What does sufficient research mean?

Sometimes it is hard to prove that you have tried to solve the problem yourself. If you say "I googled for 30 minutes and found nothing", I am inclined to believe and consider the research sufficient, unless the question really is way too elementary. If someone asks "what is a cat in Latin?" and claims that they found no translation at all, I would comment to give a link to some resources (maybe here) and close the question for insufficient research. It can be reopened later if the user tells what they have found and indicates what more they want to know.

I prefer that we leave the guideline vague and judge case by case. Like any guideline, requiring research is just a guideline. I prefer to leave room for judgement. There are great questions that utterly fail an on-topic criterion or two, and we should not close such things.

Telling users what to do

If someone asks a bad question, we should tell them how to improve it. In addition, if the question is too bad to keep, we should close it. If a question is bad enough to merit closure, we should make it clear what is needed to improve it.

The closing reasons "insufficient research" and "unclear" are close to each other, and low quality questions are likely to show symptoms of both. However, if someone is asking a pure translation question with no indication of research effort, I would consider it very misleading to close the question as "unclear". There have been questions that I would have liked to close for insufficient research although the objective was pretty clear.

Closure is always a result of overall judgement, not going through a list of a list of criteria. If the overall judgement says "close!", then the closing reason should indicate what the key problem is. Commenting is a good supplement, but the closure banner should be informative in itself.

One important thing to communicate with the closing banner is that the question can be reopened. We can close more new users' questions than we do currently, provided that we make it abundantly clear that the questions can be brought back to life once they are improved. That will also serve users: the answers will be more useful for them once they show their own thoughts. This is related to the con point listed in the question.

What resources must users check?

As I said before, I prefer not to carve rules in stone, but we should tell users what kinds of resources they are expected to check. I think the resources listed in the meta question about translation and homework questions are certainly enough.

With translation questions I would consider it sufficient to check any of the dictionaries mentioned in our dictionary list.

The relevant resources depend on the kind of question asked, and the expectation of research should (in my view) cover all question types. Lack of research has been the biggest problem in translation questions, so we should focus there.

What to do?

As mentioned, I suggest creating a new closing reason for insufficient research. If we decide to do that, the proper wording should be discussed in a separate meta question. I hope that the requirement for research is enforced. The suitable level of enforcement is left for everyone with sufficient reputation to decide — we, the users, enforce any policy we might have.