See this exchange in the comments, about closing the question.

I would like to see, as a possible reason to close a question, a judgment as to whether the OP made a minimal effort to find the answer.

I think that a question should mention what research the OP undertook to try to find the answer, before asking here.

This judgment is relative to the question and the perceived knowledge level of the user, of course. A user who seems to know the basics of how to get around in Lisp or Emacs should be expected to make a minimal effort to ask Emacs (e.g. C-h f or C-h v), IMO.

A complete newbie would not be expected to know how to ask Emacs in even a basic way. The bar could be much lower for an obvious newbie.

But the page about how to pose questions here could also mention some basics about how to ask Emacs - that would be helpful, I think. There is no shame in ignorance, and obviously, all levels of Emacs users are welcome to pose all kinds of question.

But I do think that a lack of specifying in the question just what the OP has tried should be grounds for voting to close. This is similar to what is done on StackOverflow and english.stackexchange.com. And to me, it constitutes part of asking a good question.

This is about adding this as a reason to close. That's all. Closure requires judgment and votes by multiple people.

Just one opinion.

  • 1
    Also relevant: How much background research before posting a question?.
    – Dan Mod
    Feb 26, 2015 at 18:11
  • @Dan: Yes, thanks for that pointer, Dan.
    – Drew
    Feb 26, 2015 at 19:14
  • 5
    I must admit that my tolerance for an apparent lack of research effort is considerably less if I've seen several questions from the same user which could have been answered with minimal research, and the user has previously been shown how to use the documentation in Emacs. There comes a point at which one tends to feel that it really is "lack of research effort", rather than ignorance. (Bizarre questions like emacs.stackexchange.com/q/9567/454 don't help, either.)
    – phils
    Feb 26, 2015 at 20:22
  • @phils - I agree with what you say. It's a judgment call, and that can be influenced by past behavior.
    – Drew
    Feb 27, 2015 at 16:33
  • 1
    Yes, such questions are annoying and the site gives you a way to deal with them - the down-vote. Whats the problem with that? While SO has a similar off-topic reason, it applies to questions asking for debugging help without showing effort. This one, IMO, doesn't fall into the same category. That said, it would be great if we could have all the question closing reasons that SO does.
    – Pradhan
    Feb 28, 2015 at 19:05
  • 1
    @Pradhan: As itjseyd says here: "This would enable us to close these types of questions against the CW [Community Wiki] posts, and make it possible to point novice users to on-site resources that are under our control". And: "I wouldn't go as far as saying we should downvote [such questions]". IOW, provide CW how-to-ask-Emacs and other basic info posts, and point to those when closing such questions, rather than proliferating such questions with downvotes. That is exactly what I suggest.
    – Drew
    Feb 28, 2015 at 19:21

5 Answers 5


Oh, it's the discussions like this that made me unregister from StackOverflow flagship site. Come on people, sorry for the rough language, but don't be anal about it. I've been a moderator for several years at another programming forum, and I believe that questions like the one being discussed have their own merits: they provide an opportunity for people with less extensive knowledge of the subject to answer.

I really dislike the idea of punishing for not seeing / displaying enough research by preventing answers to the question. The site as a whole benefits from answering basic questions (which could have been solved by reading the manual). In fact, if you look at the highest scores in almost any programming language on SO, you'll see that it's mostly the trivial questions, (POV-questions aside), the questions with in-depth answers and complicated subject matters don't attract as much traffic. So, apparently, closing such questions will do disservice to the site. Some times it feels unfair that you had to do more work to figure the answer yourself, while someone will get it "for free", but I think that the proper reaction is to ignore the question, if that's what you feel. Closing it will have no pedagogical effect.

Here, just to show that I'm not dreaming it: these are the highest rated questions in Python tag of SO:

enter image description here

I assure you, any of these can be easily answered through reading online documentation and loads of books, blogs and manuals that are immediately found through any search engine.

  • 3
    FWIW, I answer lots of easy-to-find-the-answer questions. I agree that people should be allowed to ask & answer them. I agree that help from others is relative to whether someone repeatedly makes little effort. I still think that SO and english.stackexchange.com have the right approach by encouraging people to make an initial effort, and to describe to others what they have already tried. And I think that lack of such info can be a reason for closing. And yes, closing questions, like downvoting them, can improve question quality. And encouraging users to ask Emacs first helps them, IMO.
    – Drew
    Feb 26, 2015 at 22:27
  • 1
    There's an anecdote re' Jewish law at the time of the late Roman empire. There was a rule demanding that all debts will be forgiven once every seven years. One of the local leaders, Hillel, suggested that this law be abolished and that the debtors will be never forgiven their debts. On the surface of it, the solution doesn't favor the debtors, however, it appears that the situation at that time was such, that the creditors, afraid of never being paid back simply refused the loans. So, the new law was actually in favor of debtors.
    – wvxvw
    Feb 27, 2015 at 22:01
  • 2
    @Drew I remembered this anecdote because I don't think that the most straight forward way to achieve any kind of justice is necessary the most efficient one. In other words: maybe it sounds just to close such questions, but it's not helpful neither for asking side nor for those wanting to answer.
    – wvxvw
    Feb 27, 2015 at 22:04
  • Dunno what to say to that; sorry. To me, these things are not black & white. There is no simple rule, whether "most straightforward way" or another, that one can rely on to always find the best outcome. It's about judgment. And it helps to apply the judgments of more than one.
    – Drew
    Feb 27, 2015 at 23:16

"FWIW, I answer lots of easy-to-find-the-answer questions. I agree that people should be allowed to ask & answer them. I agree that help from others is relative to whether someone repeatedly makes little effort. I still think that SO and english.stackexchange.com have the right approach by encouraging people to make an initial effort, and to describe to others what they have already tried. And I think that lack of such info can be a reason for closing. And yes, closing questions, like downvoting them, can improve question quality. And encouraging users to ask Emacs first helps them, IMO."

I agree with all of these points except for the two sentences about closing. A closed question is a question not wanted on site (duplicates aside). If someone asks a question which they find not trivial, but it gets closed because of not enough research, it says "we do not want ignorant questions like yours on our site". Downvoting says that "the question should be improved", this is not the same thing.[1]

This is fundamentally different from duplicates : when we close a question for duplicate, we do close it (but not in a "we don't want your question here" way, in a "let's make this question a path to the good answers" way) and we explain to the user that "this question has been asked before, it has a lot of good answers already, so he will be better helped by reading this other question than by waiting for a quick answer here". The implied criticism of "not enough research" is probably perceived, but it is still implied.

Let's not forget that ultimately, all stackexchange does is rephrasing documentation, over and over. That does not mean that the questions or answers are trivial. Sometimes documentation is not easily available. Sometimes it takes months of work to deeply understand its meaning. What answerers are doing is not copy/pasting routinely from manuals, it is using their experience and knowledge of the topic to find the very relevant passage, and then rephrase it so that the OP (and future visitors) get enough understanding to solve the problem from the question.

Just a simple test: you are looking for how to do something trivial in a language you are not too familiar with, so you ask <search engine>. In the results, you have a stackexchange page, and a documentation page. Which one do you read first?

And last, in my opinion the situation also depends on the age of the site: for older sites, such as SO or english.se, it is likely that most easy interesting questions are already asked, and that a user asking a new one should point out how his question is not already answered. This policy helps the user not waste time carefully asking a question which was already answered. But our site is still young, a lot of easy interesting questions (yes, I put the one that you mention in this category) are still available, and they should exist.

[1] Questions can be closed, editted (by the OP or someone else), and then reopened. But in my opinion, "not enough research" as a close reason does not leave much room for improvement beyond deletion. If the question is unclear, this is an entirely different issue.

  • FWIW: Questions that are closed can be reopened. It is often the case that a question is closed because it is unclear so far, and a comment is entered to that effect, then the OP edits the question to improve it, and the question is reevaluated and reopened. I do agree that downvoting is a better way of indicating problems, in general (and of course it obviates the maintenance of closing, reviewing, reopening). And the key help is adding a comment to indicate what's needed - or link to a page that provides info (e.g. about the kinds of questions to ask and how to ask a good question).
    – Drew
    Feb 27, 2015 at 15:13
  • How could the question about setq be improved, in your opinion?
    – T. Verron
    Feb 27, 2015 at 15:26
  • As I said, I think that that question is not a question about setq. I'm guessing that the real question the OP has is perhaps about Lisp evaluation or quote or function invocation - or perhaps LISP1 vs LISP2 or function cells vs value cells or symbols in general. To me, the OP should take the feedback given so far and, if there is still a question at this point, pose it clearly in whatever terms are most useful. I do not at all see the issues discussed in that Q & A as being about setq. But it's possible I'm mistaken, of course.
    – Drew
    Feb 27, 2015 at 15:47
  • Oh, and if I felt that I knew just what the real question is then I might well edit it myself, to help clarify things. I don't feel I know what the confusion/question is well enough to do that in this case. Which itself might be saying something.
    – Drew
    Feb 27, 2015 at 15:48
  • To me, what you describe sounds more like an instance of the XY problem (which are tricky to deal with too) than a "not enough research" question.
    – T. Verron
    Feb 27, 2015 at 15:49
  • About your second comment, I guess I can understand that. Still, since the question has an accepted answer, "unclear what you're asking" seems to not be unanimous. Maybe based on that answer, we can suggest a better formulation of the question?
    – T. Verron
    Feb 27, 2015 at 15:50
  • Yes, that's exactly what I think. I don't know what the real question/confusion is, but I'm pretty sure that it is not about setq. As it stands now, as a question about setq, I think it should be closed (or replaced by the real question).
    – Drew
    Feb 27, 2015 at 15:51
  • I agree that reviews of the question are not unanimous in their interpretation of the question, let alone in their proposed remedies. That in itself might be a sign that something might not be entirely clear. As for the acceptance of an answer, what can be said about it? Apparently the OP feels that the answer helped, which is good. Maybe the answer saw past X to Y. Still, to help more users it would be good to clarify/rewrite the question to be about Y, not X. (I might have X and Y reversed. ;-))
    – Drew
    Feb 27, 2015 at 15:54
  • I still do not understand why this question should be closed: maybe it is not clear to you or me, but it already has two upvoted answers, one accepted. Closing does also say to the answerers "you shouldn't have answered this", somehow... Yes the question could and should be improved. But what good would closing do that, say, downvoting wouldn't?
    – T. Verron
    Feb 27, 2015 at 15:57
  • What can I say. I voted to close it, and I gave my reasons. Others are free to disagree with my close vote. And it hasn't been closed. And it hasn't (yet) been improved. I would change my vote (FWIW) if I thought the question was different from asking what setq does, and was clear, and thus helped readers. The answers are clear, and helpful - I agree with you about that. The combination of question and answers might not be so helpful (including for people who search for what the answers are really about). It might even be harmful/confusing. This is a judgment call - needs 4 more close votes.
    – Drew
    Feb 27, 2015 at 16:05
  • I will add that this seems to be turning into more a discussion about my judgment in voting to close that particular question than a discussion of the meta question I raised here. It should be possible to think that I was wrong in my judgment about that question and yet think that the question here deserves discussion.
    – Drew
    Feb 27, 2015 at 16:20
  • I agree, and the answer we are commenting on is trying to answer the question in general (btw I just editted it to address your first comment, which was still general too). Also, I am not judging your vote in particular, just trying to understand it. It is a young website, this is a good time to establish good policies on accepted content, and so your question is a good one.
    – T. Verron
    Feb 27, 2015 at 16:41


First, "minimal effort" is vague enough to vary from person to person. An answer I could find in ten minutes might take a complete noob an hour. It might take someone else two minutes.

Second, what is a great answer for a more experienced user is clear as mud to a new user. Repeating that "page 1 of the manual says you just need to flibber the widget, which would solve your problem if you'd taken the effort to read it," is a lot less helpful than explaining what those terms mean, (and perhaps mentioning where to find it).

Third, even if the user asking the question put in zero effort, at some point somebody is going to have the same question. If the answer's already there, it's that much easier for them.

In the end, this is a question and answer site. If the answers all come down to some vague "you have the answer - if you search long enough."

  • 4
    The outcome of a "minimal effort" certainly varies, but if the question asker indicates what they've tried so far, it (a) shows they've tried, and (b) helps people to target their answers. Expecting a minimal effort doesn't mean we hold all people to the same standards of knowledge and understanding. It's about making the question as useful as possible, so that the answers can be as useful as possible.
    – phils
    Mar 1, 2015 at 19:41
  • I would agree with both your points, especially (b). But, I think the overall usefulness of the site will still be improved by answering questions that don't meet both criteria.
    – user2699
    Mar 2, 2015 at 22:35

Negative, for the following reason:

I'm more concerned with how well a question is phrased and its relevance. A question matching these conditions will be useful. For example, I asked the following question on SO:


The resulting discussion raised some issues that I was unaware of, even though I supplied my own answer (a legitimate SO activity).

Additionally, sometimes a question is raised because an problem has occurred through a third party dependency. The OP has no direct interest in the subject matter, but simply wants to fix an unanticipated error.


In your question, you make a fundamental assumption that is incorrect. Your reasoning hinges on the premise that the intent of the asker is germane. This is wrong for two reasons:

  • It is impossible to know the asker's intent. All you can know is what the asker usually wrote. Sometimes this makes it apparent whether the asker expended effort into searching for an answer, sometimes not.
  • Answers on Stack Exchange are not just intended for the asker, but for future visitors who will keep coming over the years. Even if the asker was incredibly lazy, what really matters is whether the answers are useful to future visitors.

Closing a question affects all future visitors identically, so it doesn't make sense for a close reason to depend on the intent or experience of the original poster.

By all means, if the solution can easily be found in the documentation, explain how to find the solution using Emacs's built-in documentation search tools. (But don't forget to actually answer the question.)

There is a site feature to indicate that you wish the asker had spent more effort researching the question:

This question does not show any research effort; it is unclear or not useful

This feature is downvoting. Do not use a close vote in lieu of a downvote, they have different purposes.

There was a similar debate about Stack Overflow a while ago. You can read my thoughts here and here, and Shog9's conclusion. In a nutshell, Stack Overflow had a close reason that was interpreted as “lack of effort”; it was causing a lot of harm (especially as a lot of people ended up considering that any effort that fell short of finding a solution was insufficient) and was eventually reworded to avoid this interpretation (at least in good faith).

  • Nowhere in my proposal did I say anything like what you claim I say and claim I assume. Nothing about the OP's intent or laziness or any of that. I mentioned "minimal effort to find the answer", and describing the "research the OP undertook to try to find the answer". This is to be better able to help. I said that an individual vote to close is a judgment "relative to the question and the perceived knowledge level of the user". I made it clear that this is not about any non-relative entry level - all questions at all levels are welcome. Nothing to do with awarding "A's for effort".
    – Drew
    Mar 16, 2015 at 1:24
  • This is an Emacs site. IMHO, a major part of helping Emacs users here is helping them learn how to ask Emacs. This is nothing like telling someone coldly to "RTFM!" - it is in fact the opposite. Emacs is different: learning it is very much about learning to ask it about itself - at all levels, not just the doc. Learning to dig into Emacs is learning Emacs. That is the motivation of my suggestion - it has nothing to do with punishing lack of effort or rewarding evidence of effort. Nothing at all.
    – Drew
    Mar 16, 2015 at 1:29
  • 1
    @Drew Your reasoning does assume that the question is for the asker. It completely falls down when you consider that the question is for future visitors. Mar 16, 2015 at 1:40
  • Not at all. And you've shown no evidence or argument supporting that. You seem to making a lot of assumptions. For me it is very much about questions and answers (both) being for all readers. That's in fact the real payoff (here) of a help-users-learn-to-ask-Emacs strategy, which is the motivation here. It is the reason for my encouraging questions and answers that lead in that direction. Nothing could be farther from the truth than to think that I am focused only on the particular OP or only on the OP's question or only on the OP's formulation of the question.
    – Drew
    Mar 16, 2015 at 2:35

You must log in to answer this question.

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