There is a huge number of questions using the term "do it for me" question (such as [1], [2], [3] and many more), and I would like to understand what the term means. The most precise definition I have come across has kindly been provided by Marijn in a comment, and says:

  1. asks to draw something based on an image or a short description ("I want to draw a normal distribution"),
  2. does not provide any code or evidence of research or effort,
  3. the drawing that is asked is complex, and
  4. it is very specific and therefore probably not useful for anybody other than the OP

(but 5. it may serve as a good showcase for techniques used in the answer that are useful for other people).

My main question is very simple:

is this the meaning most of the users agree on?

That is, if you have a (very) different definition in mind, please give it in an answer.

My subquestions are:

  1. Are "do it for me" questions necessarily drawings (and not tables or equations, say)?
  2. What if the OP provides some code that can hardly be used to answer the question?
  3. How is complexity precisely defined? That is, what if the OP asks for a complex output, which is however rather easy to achieve with some trick or package?
  4. If a question is very narrow and the answer is unlikely to benefit anyone but the one who asked the question, do we have a separate way of referring to them (e.g. "low general benefit"), or even close them?

Any answer is welcome. In particular if it also explains why the "do it for me" term is used. Naively in any question the asker asks the community to do something for them, namely answering the question. (Which is what I am doing here, too. ;-) )

For the linguists, I am interested in the intensional definition of the term "do it for me" question.

If otherwise useless this thread can be interesting to exemplify a clash of cultures between people from different fields. There seems to be a rather large number of users who prefer a vague qualitative approach over a precise one. Maybe language is inherently like this. On the other hand, mathematics is communicated in the usual languages, and at least in this context is possible to make precise, falsifiable statements.

  • Your Wikipedia link has little to do with regular language use. As it says explicitly, this is a term used in mathematics and logic.
    – Alan Munn
    Commented May 7, 2020 at 1:21
  • @AlanMunn I just feel that you, for one reason, or others, do not like this question and .... This is not certainly the first question which asks the community to explain (or define or however you want to call it) a term for the asker. And this is what I am doing here. I am also doing this for a reason: the emotions that come up in the context of "do it for me" vary a lot between users, and I strongly suspect that the types of questions associated with this term strongly vary among the users. I cannot see anything wrong with trying to understand how others interpret it.
    – user194703
    Commented May 7, 2020 at 1:25
  • 3
    I'm not telling you you can't write questions. You've been given various fairly convergent definitions by different users about what constitutes a "just-do-it-for-me" question. But for various reasons, you don't really like those definitions, because they fail to be sufficiently precise, and you demand precision. And I'm saying that precision is not how regular language use works. This is not to say that people don't agree on things, but they agree operationally because a sufficient number of people implicitly "agree" to the application of the term to individual items.
    – Alan Munn
    Commented May 7, 2020 at 1:39
  • @AlanMunn And my concern is that this precisely did not work for the term "do it for me" question because from the context this term is used, different users associate different types of questions with that. You are saying that everyone agrees on how this term is to be used, and I am saying this is not the case. What is wrong with trying to find out?
    – user194703
    Commented May 7, 2020 at 1:42
  • 2
    There is one category of question that hasn't been mentioned, but which I think is the canonical example of a "do this for me" question -- a request to produce code for what is obviously a homework assignment. Some requesters have actually admitted as much. and when an identical request (usually a diagram or drawing of some sort) shows up within a few days, the purpose is clear. My reaction is to call it out; I certainly won't answer in detail. Commented May 7, 2020 at 1:56
  • @barbarabeeton Thanks, that's helpful!
    – user194703
    Commented May 7, 2020 at 2:43
  • So the upshot seems to be that usefulness of a question plays a role. Whether or not a question is useful is ironically decided by the same user(s) who would argue that they it is better if others who actually answer questions edit the titles of questions because of what? If you can judge whether or not a question is useful then you should not dump the task of editing titles, which can lead to strong reactions, on others. Do it yourself if you think that is a good idea.
    – user194703
    Commented May 8, 2020 at 17:04

4 Answers 4


Yes, the characterization of a "just do it for me" you give matches quite closely with how I used the term in my question. It also matches various other characterizations to be found in the other meta questions you link to in your question. While people seem to disagree about whether they should be answered or not, most people do seem to admit that the rough description given is what fits the category. Like any categorization in regular natural language, the category will be fuzzy, and not definable in terms of necessary and sufficient conditions.

My answers to the sub-questions:

  1. I think table and equation images posted with no code fit into the category but they have a higher likelihood of being useful to others, perhaps.

  2. Asking how much code needs to be supplied can quickly lead to the Sorites Paradox. There is no magic amount, but most people would like to see some code that shows that the person has tried something that resembles some or part of the image they're trying to reproduce.

  3. Complexity is not precisely defined. This is the nature of language.

  4. Again, there is no precise definition of what would count as very narrow, In the case of very specific diagrams, though, although there may be general techniques used in answers (one can obviously learn from any code) their usefulness seems limited.

  5. I think the term gets used because it picks out a class of questions that many users perceive to be a kind of abuse of the "social contract" of the site (and I'm specifically avoiding using the term "rules" here.) Namely, some measure of prior research is expected of questions, and these sorts of questions very clearly don't show any at all. This combined with the relative amount of work it might take someone to answer the question is why some think that they are an abuse.

But I also think that most people have also realized that there will always be people who find it fun to answer these questions and therefore there's nothing to be gained by trying to legislate against them (something I agree with).

  • 3
    Helpful, so +1.
    – user194703
    Commented May 7, 2020 at 3:42
  • 2
    I didn't say the question was unanswerable... :)
    – Alan Munn
    Commented May 7, 2020 at 3:42

My interpretation of a do-it-for-me question follows your collection of points, but with less precision. I'd almost define it more broadly with the following characteristics:

  1. Asks to create something based on an image or a short description.

  2. Does not provide any code or evidence of research or effort.

This therefore includes things like drawings, equations, tables, layouts and the like. This even includes someone asking "What is that symbol used in X to represent Y again?" Why? For many reasons, depending on your interpretation of the question (because yes, they're using their language, which may not mean the same thing for everyone). Here are some interpretations:

  • If it's just looking for the name of the symbol, then it's off-topic here, so move along. Off-topic questions should be closed.

  • If they know what the symbol looks like and only show you an image, then they can do a bit of research via Detexify or the Comprehensive LaTeX Symbol List to search for the symbol. From there, if they didn't find anything, perhaps they can include an image of their drawing, show their search on Detexify and mention that they've covered the symbol list, but without that an observer can argue that this person is leaving all of the research on the community's shoulders.

In short, and this echoes Alan's mention of language, its usage and interpretation or meaning, there would only exist rough guidelines for what the general audience would consider a do-it-for-me-type question. It's not a list of definitive check-boxes, nor is it something that necessarily qualifies when it meets all the criteria. Go figure! We've seen numerous examples of this on the site as well, where a question is posed without much research effort, yet it's received acclaim beyond the average. Examples include (literally just taken from the list of questions sorted by vote):

When you look at any of the above question, they seem to meet both criteria listed above (create something from a short description and doesn't provide much in terms of research). Some of these may have stemmed from a conversation in chat, or on Meta, about a canonical question related to some set of FAQs. That is not always evident from the post and therefore the do-it-for-me phrase here may have a different meaning. Others may be such generic or fundamental questions that there's no way around asking the question in that way.

Since the interpretation is inherently vague, it varies in how its handled from person to person. Some feel that the questioner may be a novice and should be guided or hand-held through the process from start to finish. Others feel that such posts should be dismissed as it doesn't deserve their attention. We can't fit everything into a particular set of rules or boxes, because invariably you'll one day find something that should fit the box, but doesn't... or fits a particular box but doesn't have the right colour... or fits a particular box but doesn't have the right shape... or...

My answers to your sub-questions are:

  1. Are "do it for me" questions necessarily drawings (and not tables or equations, say)?

No. There are a broad range of questions I feel that are not drawings that ask to create something.

  1. What if the OP provides some code that can hardly be used to answer the question?

That's okay, if it shows some form of research or effort. Why not?

  1. How is complexity precisely defined? That is, what if the OP asks for a complex output, which is however rather easy to achieve with some trick or package?

Do-it-for-me questions have nothing to do with complexity in my opinion.

  1. If a question is very narrow and the answer is unlikely to benefit anyone but the one who asked the question, do we have a separate way of referring to them (e.g. "low general benefit"), or even close them?

Scope of the question has nothing to do with do-it-for-me type questions. The question can be very narrow and have only a singular use/benefit, and that's fine. If it shows that the user has done some research and effort in trying to achieve their goal, then that's good.

  • 1
    Thanks for bringing up these highly voted posts! They are among the items which confuse me most. It seems that if some novice user asks a rather elementary question with not too much research effort, the outcome is completely unpredictable. Either they hit the jack pot and land something like this question and get votes beyond imagination, or they get a lot of comments in a strong language and their question gets closed. Or something in between. I could imagine that some find this irritating.
    – user194703
    Commented May 7, 2020 at 18:10
  • 1
    BTW, as for 2. It happens on a regular basis that an OP copies a random code from somewhere just to have an MWE, regardless whether it really relates to the question. I find this more confusing than not posting an MWE because if there is a code I'd typically assume that the OP sort of understands this code and I do not have to explain these things when answering.
    – user194703
    Commented May 7, 2020 at 18:13
  • @Schrödinger'scat: Yes. In the early stages of the development for the site, elementary questions like those lay the foundation for knowledge. Nowadays such questions may have varying levels of "success," unless they contribute more broadly. And then again, maybe not.
    – Werner Mod
    Commented May 7, 2020 at 18:14
  • 1
    The thing I still do not understand is why one needs terms like "do it for me", which is clearly a negative term. No one is forced to even look at these questions. Some are more excited about one type of questions, some about another, and one can always express the opinion in a vote. Why is it that some think they need to go beyond this and use such terms on questions? If one thinks a question isn't interesting, one can just ignore it. (I do certainly see why someone could say we should ban homework questions but this is not what this is about.)
    – user194703
    Commented May 7, 2020 at 18:20
  • 3
    @Schrödinger'scat: Perhaps "do it for me" is a shorter version of "I don't want to do any research and figure this out myself, nor do I want to show that I've done anything to explore the possibilities, so you - the answerer - do it for me, please and thank you."
    – Werner Mod
    Commented May 7, 2020 at 18:46
  • 1
    I think there's a fundamental difference between the highly voted questions you point out here and those that meet the criteria you give. You're using 'create' is two different ways. In the question you link to, the "product", i.e., the thing created is the answer itself, not the result of running the code in the answer. This is certainly true of examples 1, 2 and 4, and even to some extent 3. These are general questions the answers to which have value to many people. Their value lies in the answer. So lack of research can't be equated to lack of value.
    – Alan Munn
    Commented May 7, 2020 at 19:16
  • ... cont. But in requests for particular drawings, the product is the drawing, not the answer itself. Of course the answer may have some value to the site, especially if the code is good/interesting/well explained. But ultimately these questions aren't about producing an answer as a product, i.e., a contribution to the site, but a contribution to the particular document the OP is writing at the moment. That's why I think they are fundamentally different types of "creation".
    – Alan Munn
    Commented May 7, 2020 at 19:18
  • I have a feeling that I have responded to more than my share of questions that didn't show any code; if the symptom described is puzzling, but is something I've seen before, I will sometimes give a quick guess in a comment, but at the same time ask for more information, and a compilable example. I figure that if I can send the OP off in a possibly fruitful direction, a useful MWE might come along more quickly than if the OP just continues to flounder. (But that's not, strictly speaking, an answer, although it often results in the response "Oh, that's it.") Commented May 7, 2020 at 19:36
  • 1
    @AlanMunn "So lack of research can't be equated to lack of value." Finally we have found something we seem to agree on. While I definitely see that one could find it impolite to shoot a question without any research, the resulting thread may still be useful for many. Sometimes even packages result from low-research questions.
    – user194703
    Commented May 7, 2020 at 20:00
  • @AlanMunn: Valid point(s)...
    – Werner Mod
    Commented May 7, 2020 at 20:56
  • 1
    Of the cited high-vote questions, three of them have question id numbers less than 1000 so they can be considered "foundation" questions. There are very few topics of this level that haven't been addressed already. Commented May 10, 2020 at 23:44
  • What's the protocol for trying to identify a symbol or font that isn't in the Comprehensive Symbol list, isn't in the LaTeX Font Catalog, and can't be identified with Detexify? I have one such inquiry, but am loathe to even ask lest I draw the wrath of the higher-ups or worse...lose rep. Commented Aug 23, 2020 at 17:07

As I understand it, a "do-it-for-me" question is a question in which the circumstance plays a role that the questioner has a task to do and is not so much concerned with gaining the deeper understanding necessary for accomplishing the task him-/herself.
For the questioner the "do-it-for-me" question is a means to have the task accomplished by those who answer the question, or to have "pieces" delivered by those who answer the question, which he or she can then with little effort put together into a solution to the task.

To put it casually:

With a "do-it-for-me" question the focus of the questioner is not on acquiring the knowledge he or she needs for doing the donkey-work himself/herself but the focus of the questioner is on having the donkey-work done by those who answer the question.

Since the reasons for writing "do-it-for-me" questions can be quite diverse, I am not one of those who are fundamentally against "do-it-for-me" questions—as long as the questioners deal openly with the circumstance that it is a "do-it-for-me" question. If you already know a little bit, such questions sometimes can be a nice pastime.

But I don't like questioners trying to hide the fact that there is a "do-it-for-me" question. This often leads to questions being asked in a more general way at first—in the hope that there is someone among the respondents who provides what is needed for solving the special case. If this doesn't work out, by and by the question is narrowed down to the special case: The questioner in his/her comments by and by mentions additional conditions that were not mentioned before—or the posting containing the question is changed by and by accordingly—, which makes it necessary to take a completely different approach, so that the work done in the initial answers is in vain. And those who answered arrive at the unpleasant realization that one could have saved oneself this work if the questioner from the start had said everything.

Do not confuse "do-it-for-me" questions with questions that also are motivated by attempts at getting a task accomplished, but where the problems have made the questioner aware that he or she lacks knowledge, and where the questioner now wishes to acquire this knowledge in order to be able to cope with similar tasks without assistance in the future. (That wish may be linked to the desire to get a concrete example where one is familiar with the underlying problem and therefore can more easily learn how to approach such things.)

Classifying a question as a "do-it-for-me" question implies making a statement about the underlying motivations of the questioner. Identifying underlying motivations often is difficult. Cautiousness, thoroughness and conciliatoriness are needed here. Especially since TeX/LaTeX is something that for beginners is often accompanied by a steep learning curve, which means that beginners may quickly reach a point where they have to ask others for advice on the source code for documents they want/have to write with TeX/LaTeX. Furthermore, StackExchange is a communication medium which allows only very limited insight into the situation/motivations of the individual communication participants.

I like the idea of TeX-LaTeX StackExchange of having a collection of questions that explains the many individual basic facts in such a way that you can "assemble" your own solutions to your specific problems/tasks using this collection.

But I see a problem and this problem is one of the main reasons why I don't reject every "do-it-for-me" question:

Many beginners are not familiar with the programming-paradigms, concepts and terms that underlie TeX and LaTeX. It is therefore difficult for them to break a problem down into partial aspects in such a way that individual aspects of the way TeX/LaTeX works can—for the sake of re-usability—be explained separately.

This breaking down and then explaining the individual aspects, and how they work together, needs to be done for them and shown to them by someone who is familiar with the subject.

  • I really like the last paragraph (and the implications thereof which you are careful enough not to spell out ;-).
    – user194703
    Commented May 12, 2020 at 0:28
  • 2
    I kind of disagree with this. The goal of Stack Exchange is to provide a repository of useful, i.e., reusable knowledge. A do-it-for-me question, which has by itself limited reusefulness, could be split into multiple reusable questions (or maybe reduced to just one question about one aspect of the drawing). If the original question is answered, then the user has no incentive of asking the specific questions, so the specific reusable knowledge does not appear on the site. This in turn causes other users to ask do-it-for-me questions, because a. they can't find the reusable knowledge that they
    – Marijn
    Commented May 12, 2020 at 15:58
  • 2
    need on the site, and b. they see that do-it-for-me questions are acceptable. So it is a negative loop that leads away from the goal of the site.
    – Marijn
    Commented May 12, 2020 at 16:00
  • @Marijn I do not know how you know what the aim of this site is. It would be great if you were right, in which case many things would have to get fixed, and the problem of questions of low applicability is IMHO (by far) not the most urgent one. However, to the best of my knowledge this is a site which is run by a company, and the goal is the goal of any company, namely making money. This is (perhaps sadly, certainly sad to me) not academia. If it was, someone would do something about the outrageous level of plagiarism.
    – user194703
    Commented May 13, 2020 at 21:27
  • @Schrödinger'scat the founder of Stack Overflow has stated this many times, for example twitter.com/codinghorror/status/991082088689381376. Related discussions: blog.codinghorror.com/…, meta.stackexchange.com/questions/217115/…, meta.stackoverflow.com/questions/351354/…, meta.stackexchange.com/questions/297483/…
    – Marijn
    Commented May 14, 2020 at 5:49
  • meta.stackoverflow.com/questions/288133/…
    – Marijn
    Commented May 14, 2020 at 5:49
  • @Marijn Huh? This is a bit a different question, isn't it? (Ironically this question is about the definition of "do it for me" question, and when I asked you, you gave me a rather nice definition. Now it seems that this has been a bit shifted. But of course you are very welcome to cast this interpretation in an answer.) Also under your links there is this post. Whose answer should we take to define our opinion? They seem similarly divided as we.
    – user194703
    Commented May 14, 2020 at 6:04
  • @Schrödinger'scat you asked me how I know what the goal of the site is. I replied that the founder of Stack Exchange has stated this goal. Then I linked to several questions where people discuss this goal. It is related to do-it-for-me questions because in my opinion, and as mentioned in those discussions, do-it-for-me questions (or negatively called gimme-teh-codez) are contrary to this goal.
    – Marijn
    Commented May 14, 2020 at 6:09
  • But you are right, other people share your opinion that answering such questions is ok. That's why it is called a discussion :)
    – Marijn
    Commented May 14, 2020 at 6:10
  • @Marijn Well, I have seen so many posts that mutually contradict each other, and we can now start a competition in digging them out. You have an opinion, I seem to have a somewhat different opinion, and we can just agree that there is no perfect agreement.
    – user194703
    Commented May 14, 2020 at 6:18

I still do not really know what a "do it for me" question is.

However, from how the term gets used I see that this is a term that, willingly or unwillingly, divides the users. Ultimately, by using such categories there will be (or already is) a division between "good" users who ask useful questions and perhaps not so good users who ask (or even answer) "do it for me" questions.

Some seem to look down on those who ask questions that one can instantly answer. I, too, would really love to be in a community in which everyone is somewhere close to my point of the learning curve at which I am, certainly this would help me most. It may just not be realistic to wish this to happen. There will always be some who know stuff better than I, and if I ask them something what they think is trivial, I'd still appreciate it to get some answer that helps me and is tailored to my needs.2

I'd like to ask the community of users not to try to oversimplify matters, and in particular to avoid using divisive, oversimplified terms.1 Rather, if you want to judge or rate something, vote on posts. If you are not excited about a post, just do not upvote (or even downvote if you must).

Please avoid introducing terms that assess the quality of posts beyond what voting does.

1 I am aware that most users do not have the intention of dividing the users. Nonetheless the effects of using such oversimplified terms are unfortunately divisive.

2 I totally can see how a question which leaves all the efforts to the answerer may not very popular. Yet, for me personally, it is hardest to answer the question which come with an extensive preamble of unrelated stuff, which leaves the answerer with all the weeding tasks in addition to answering the question. To the best of my knowledge no one ever found a category for those. Let's keep it that way.

  • 1
    It should be noted though that this is not just about the opinion of the community about the questions or the behavior of other users. It is also about the Stack Exchange model in general. The site model explicitly defines overly broad questions as questions that should not be on the site and should be closed, with the close reason "Needs more focus". Closing prevents answering. Many do-it-for-me questions are very broad/in need of focus. So the site model itself tries to prevent such questions, it is not just the opinion of the community.
    – Marijn
    Commented May 11, 2020 at 14:22
  • @Marijn It is perfectly fine to close questions that do not belong on the sites.There exist terms for that which can be used. It is IMHO not a good idea to introduce an extra term that has the effect of dividing the users.
    – user194703
    Commented May 11, 2020 at 16:26
  • Ok, that is a fair point, I agree.
    – Marijn
    Commented May 11, 2020 at 18:00

You must log in to answer this question.