Humbly, I have 7 years experience in the network and have asked about 1,500 questions throughout different sites in it. I have recently joined Medical Sciences SE where I was managed to ask 37 questions until getting blocked from asking more for some time; it is a site in which almost any question is closed, either by lack of evidenced prior researched, asking for personal medical advice, or both. The OP is lucky if its migrated.

Many of the aforementioned closings happen to new users who didn't even know about the "draconian" demand of evidenced prior research which is perhaps unique to Medical Sciences SE (due to "evidenced").
This often brings a redundant chain of comments between very experienced users (including mods) to these new users, which could otherwise be prevented and time would have been saved for all involved.

From my experience in Chemistry SE, Biology SE and Psychology, Neuroscience SE, Computer Science SE, Super User SE and Unix and Linux SE, a similar problem happens there as well but not as frequent as in Medical Sciences SE.

My suggestion

I suggest to consider adding a unified structure for asking questions to all STEM network sites, so that all such websites (such as Medical Sciences SE) will have a unified structure for associated questions such as this one:

Background data which includes an explanation about how this session deals with general audience and isn't a personal medical advice request or an homework solution.

Evidence of prior research

  • ...
  • ...

My problem

A problem in understanding one or more sources in evidenced prior research or of experiencing a conflict between one or more sources to one or more sources.

My question

A question based on the problem in understanding the condition (identical or similar to a question in title --- if there was one).


  • Isolating a problem based on prior research can prevent what is very often assumed as XY problem
  • Clearly requesting evidenced prior research will raise the change for mutual learning, will ensure good sources, will ensure more content and better SEO (if all goes right, hopefully) and will on the way prevent heresays
  • It should prevent massive amount of closings as common in life sciences oriented network sites as I have clued as well as subsequent miscommunications between very experienced users (including mods) and new users; thus it will allow the mods more time to deal with more important stuff such as COC violations, migration, merging and alike


Per my opinion, evidenced prior research and isolated problem to solve in understanding, aren't the perfect path for a good question but they can certainly lead in that direction.

One could ask how to "access" the usage of such ready injected template to new users and what to do with old questions.


5 Answers 5


I agree with the answer of Rebecca J. Stones that not all questions fit nicely into a template.

However, this issue depends on the compelxity of a template. If a template asks to show which error message you get from your code, while you may not get an error message (but just wrong output), or your question is not about code at all, then the template does not fit. But if the template is simple and general, such as the one proposed here, then it will most likely fit many questions. If the second and third template item are merged (leaving just Research and Question) then it will fit even better.

SE needs to do research to find out if the fit is good enough, of course, similar to process of evaluating the Ask Question Wizard on Stack Overflow.

Of course it should be possible to not use the template, for example by clicking a button that says "I want to ask a free-form question". Maybe enable such a button as a privilege for, e.g., 200+ reputation, but that is a different discussion. But when the template is the default, then people are at least made aware (more) of what is expected of them - to do prior research.

The awareness is important aspect of this issue I think. The change is not so radical if you look at the directions for asking a question provided next to the question input field. On Medical Sciences:

enter image description here

Cf. Stack Overflow:

enter image description here

So, the requirements are already there (although the wording could be improved). Incorporating these in a template would therefore not be a big change conceptually, it would just reinforce the guidance that is already provided. Many question askers don't read this guidance, or don't think it is important, or don't know how to apply this to their question - a template could help to bring this to the attention of the asker and help them structure their thoughts about the question.

In my opinion, every change that has the potential to improve question quality should be implemented, especially since there is a clear problem to be solved, such as on Medical Sciences where according to the OP almost every question is closed. The template as proposed could warn the user, for example if the Research section is left empy, or if it does not contain any links, with a message that says something like "it appears you did not provide prior research, this could mean that your question is not a good fit for this site, and that you may not receive an answer/your question could be closed/you can expect to be asked for the research in a comment". One of the biggest problems of SE, network-wide, is managing expectations from askers. I think that a template such as the one proposed in this question can be a step towards managing these expectations.


This is a radical change, which will affect almost all of Stack Exchange (think about which sites are STEM sites; the largest non-STEM site seems to be English Language Learners), i.e., it would affect millions of users every single day. Thus, even if it were a good idea, it needs far more evidence to support it---a feature request this massive needs hard data not "I think it's a good idea".

That aside, there are major reasons I do not support this (as proposed):

  1. I do not want to be forced to ruin the flow of my future questions in order to fit into a one-size-fits-all template. Not all questions fit easily into a template.

  2. The template, as it stands, feels like it would be difficult to understand for new users.

    • I don't think it's clear that "evidence of prior research" applies to the OP, and includes things like "Google it" (as opposed to evidence that e.g. a scientist has previously researched the topic).

    • I don't think "my problem" and "my question" are always easily separable. E.g. sometimes the problem is simply "I don't know the answer to my question". [I'm also not so keen on "my question"---we're a community, and work together.]

  3. I'm fairly sure people are just going to ignore it, ask their question in one field, and type nonsense into others fields and click submit. The same low-effort questions will be closed for the same reason: for being low-effort.

  4. I'm fairly sure the template will need to vary from site to site. Templates suitable for medicine are not going to be suitable for mathematics. (Where do you even copy/paste the source code?)

  5. In my opinion, having every question follow an identical format just looks ugly.

If the proposal was for optional site-specific question templates, then I'd be more inclined to support it. But even then, it still would need tangible evidence for me to be persuaded.

  • 1
    @JohnDoea You said "old questions could be kept as is or locked". They could, however doing so can affect how new members of the community view the site guidelines. If the question or answer does not fit the guidelines, it should be put on hold (closed) and edited to conform or removed. Feb 22, 2020 at 13:16


When I asked a question recently on graphicdesign.se I saw there a feature of SE software which I hadn't seen before (screenshot below).

It isn't exactly what you suggested but it's better than nothing at displaying some usage guidance to the user when they ask a question. Possibly, I don't know, the moderators of your sites could get such a thing added/configured even without changing the software -- that is, if what you're suggesting is a software change, like a Wizard to prompt the user for the different parts of a question.

Maybe you're not even suggesting a software change, i.e. perhaps you're only suggesting a more-formal set of well-understood conventions about what type of question is on-topic.

Skeptics.SE for example has a strong (well-defined and strictly-enforced) convention like that, defines in its FAQ on its Meta -- i.e. that questions must include a reference to a "notable claim" and answers must include one or references to authoritative/peer-reviewed literature.

BTW I also think that sites ought to have some community-authored Welcome text that's a bit more evident/discoverable than away in a Help section and/or in one of the meta-FAQs -- for example it could be:

  • On the sign-up page
  • On the Tour page
  • In the Help and as a FAQ
  • Linked from the navbar
  • An automatically displayed hyperlink when a new user posts a new question or answers
  • In a post notice (for answers) or close reason (for questions)

enter image description here


Alternatively (maybe better as it's not tag-specific) there's the mechanism by which MedicalSciences.SE displays a disclaimer (perhaps that's not configurable though and is instead baked in to the site design):

enter image description here


A third way SE can implement (has previously implemented) this is, for example, new or anonymous users are directed to a site-specific How to Ask which they must agree to before they ask.

  • 10
    What you are seeing there is a tag warning.
    – Luuklag
    Feb 19, 2020 at 9:15
  • 4
    Thank you for naming it (so see also What's the protocol for requesting a tag warning?).
    – ChrisW
    Feb 19, 2020 at 9:37
  • I could be wrong but I believe that tag warnings are not going to be useful for this purpose. The problem is not about the subject they are talking about. The problem is how the question has been researched (if at all). Prior research information provides information on what you know and don't quite understand. This then leads to more useful answers which don't repeat known information for a start. Feb 22, 2020 at 11:51
  • 1
    @ChrisRogers I think a warning that is site-wide and not tag-specific could be helpful, especially this format (a short list and a hyperlink) and this location (under the new question when it's being written).
    – ChrisW
    Feb 22, 2020 at 11:59
  • 1
    @ChrisW - That is a possible resolution. MedicalSciences.SE does have a site warning about the fact that it is for educational purposes only. The same area could be used for this purpose. Feb 22, 2020 at 12:08

A similar idea to the tag warning talked about by @ChrisW could be used for this purpose.

MedicalSciences.SE does have a site warning about the fact that it is for educational purposes only. The same area could be used for this purpose.

In the case of MedicalSciences.SE this area could have the following added:

All questions and answers require citations to reputable sources in order to back any claims made regardless of background and provide evidence of prior research.

The thing is that this was discussed here at Health Stack Exchange has rules that violate the general rules when MedicalSciences.SE was Health.SE and although there isn't a SE wide solution as of yet, MedicalSciences.SE has come up with a solution where there are standard comments suggested in the meta site for MedicalSciences.SE

It is then down to the OP and new members to read the linked guidance and respect the guidelines put together by the relevant community.


Such a form can work well for bugreports, but even there you should be free to remove sections that are not relevant to the bug (e.g. summary/intended behavour/actual behaviour should often be mergen for feature requests).

In this QA format, it adds noise. The reader sees and reads the headlines, that do not add much to a nicely formatted question. So the form could be prefilled with (later invisible) comments like "please add the following ...", but you should not be forced to put your question into a 1,2,3,... format when you can structure it in a free form much better.

You must log in to answer this question.