Does Web Apps Stack Exchange Require Research?

Question

The Web Applications Stack Exchange site is in an interesting position regarding the topic of showing research has come up several times over the past years. As a community manager, I wanted to give the discussion a fresh start to help the community come to a consensus. However, if we can not reach a consensus as a community, I will discuss it with a few CMs to come to an agreement for a site policy to put it to rest.

To set the stage, I see two perspectives after reviewing past discussions on the matter:

1. Little to no research is required

   Support of this viewpoint is under the premise that question asking on web apps is to seek help first and that imposing research requirements might discourage users from asking questions. Closing questions should be based on their lack of context and clarity, rather than a lack of research. They believe that not requiring research creates a welcoming environment, especially for new users, and allows for more active participation.

2. Demonstration of Research Efforts

   On the other hand, requiring research argues that providing context about what has been tried and where the user is stuck is more beneficial than simply asking a question. This not only helps in a better understanding of a question but will also prevent duplication of efforts by the answerer. Showing research can make a question more engaging and relevant while giving it better odds of receiving a response and participation from the community. Questions that do not demonstrate research effort appropriately should then be considered for closure. However, it might encourage new askers not to ask due to the higher bar for participation.

The Ask

Given the previous discussion on research in the web application community, I would like to pose the question to you all: What level of research needs to be demonstrated for a question to be acceptable?

Furthermore, should it become a requirement for users to show some level of research, like searching the help center of the web application in question before posting it? How best should that be demonstrated, and to what degree, if any, must it be established?

We will leave this open for community input for four weeks and stop considering feedback to the question on October 5th, 2023. After that point, if there is overwhelming support for one option, we will adopt that as policy. If there is not, the CM team will craft something sensible between all suggestions offered and existing network norms.

Answer 1

Downvotes, not close votes, for bad poorly researched questions

I've been on a lot of sites, and this is what I've seen work the best. And even then, downvotes are reserved for the questions that could have been solved very quickly and easily by pretty much anyone (as we're not all experts), not challenging problems that are brief only because they don't include OP explaining that they're getting nowhere fast. (On a site like this with such a wide breadth, it can be hard to tell these types of questions apart, especially when there are no answers yet. That's one of the reasons why both should stay open.)

To be clear, I think that it's usually a good idea for users to do at least a little searching before asking, especially when they're likely to be downvoted if they don't. It just doesn't need to be a requirement. (Obviously, if your question was already asked here, it will be closed as a duplicate, but that's not quite the issue at hand.)

To better understand my reasoning, let's use What do the red dashed lines mean in Google Maps? as an example since I've already spent a bunch of time on it (see also Reopen "What do the red dashed lines mean in Google Maps?").

The state of documentation

For most of the web apps I use, I have no idea where the documentation is, and the documentation that's out there is often... not good. If I can't figure something out by myself, I search the internet for the answer.

Google Maps has next to no internal documentation (that is, interacting with the website interface does not explain much—in this case there is no map key). When I try to search for the external Google Maps documentation, I instead get the documentation for developers and Google Map Platform (for businesses), both of which may look potentially relevant from the search results but aren't. It's not until I search for the help that I get a relevant hit which has the information for users of the website and apps. However, it doesn't bear fruit. Searching for different combinations of "lines" or "red" turns up nothing. Going to several of the relevant articles and expanding the accordions manually (ugh, collapsed accordions) also brings up nothing relevant. I don't think the answer is in there, but I didn't read everything (and shouldn't be expected to).

When I tried to find the answer for this via a Google search, I found conflicting results:

• What does the red Dotted line around some landmarks (beaches ) mean?: "The red dotted line represents the extent of the beach." (This is only correct in the thread it was asked in; in our example, the circle is around the entire island, not just the beach)
• What does the red dotted line mean on roads, e.g., M25 J27 now?: "Currently experiencing Moderate to Extensive delays" (This is a different red dashed line)
• "The dotted line is not directions. It just indicates that you can’t get to the destination by car so it brought you to the closest place and you’ll have to walk the rest of the way. The dotted line is not walking directions but I guess I can see the confusion." (Yet another line)
• Twitter: "Google Maps has a red dashed line around Crimea, btw -- I assume to indicate it's disputed territory." (Right line, wrong answer)
• Many more results, most of which are about other lines. (How many different types of red dashed lines are there? Am I even looking at results for the website?) I didn't see the right answer.

Now, there is one result that could have solved the problem in an instant. It's the link to the question on Webapps, which currently ranks #2 on my search. I could point out all the ways that we can be better than the other sites on the internet (ranked answers! no chatter! edits!), but since you're already this deep on Meta I'll assume you don't need to hear all that.

Research requirements as an obstacle for experts

Writing the answer to the question would have taken less time than even finding the official help for the website, much less searching through that, or writing about it, or the other searches I did. Also, to be brutally honest, I don't care about this research. It's only making it a longer scroll down to the bottom of the post to either post an answer (we all might already know the documentation sucks), or to see what the answer is (we're not here for distractions!), depending on why you're looking at the question. Hyperbole aside, the point is that it doesn't help a potential answerer get closer to the answer. And if you're the one who's expected to do all this research, you're probably looking for another site to participate on next time instead, if you don't just decide to settle for whatever wrong or suboptimal answers you can find in Google.

Even before writing all this, I've already put too much effort into reopening this one question. Looking at meta, I think I'm not alone. I also get the feeling that moderators are similarly spending a lot of time; since they are the ones who close the questions, they're the ones who have to answer inquiries about how to reopen them from all directions, including meta. We could use the time that we're currently spending closing and reopening these questions on other matters, like answering questions.

Let the experts fill in the gaps

I think it's also relevant to discuss questions that don't include all the relevant details to solve. It's only natural; people who ask questions typically don't know the answer, so they often won't know what additional details to add or factors to consider. It might be tempting to close some or most of these questions as needing more details and barrage them in the comments asking about various minutiae. However, that's not the only option, and in most cases it's not the best option.

An expert knows the caveats of different approaches. Instead of asking a lot of questions (e.g., "Mac or Windows?"), experts give thorough, wiki-like answers that let the reader know the pros & cons and decide what solution works best for them, and also work no matter what setup the reader has ("This way on Mac, a different way on Windows"). And, if all goes right, they help people other than the person who asked the question.

Answer 2

No, we shouldn't make research mandatory.

Showing research effort is a good practice that we should promote among the community, but I think we shouldn't close questions that don’t show research. This would lead to the closure of most questions on this site, which would further reduce the participation and activity of our already small community.

Most of the questions on this site do not show research, as you can see by browsing through them. If the question is clear and understandable to anyone who uses the web app in question, then it should not be closed as "requires research" via the "needs details or clarity" close reason or any other close reason.

Here are some examples of questions that were closed for not showing research:

• How do I delete my Facebook account? — This was the very first Web Apps SE question, posted 13 years ago. It was closed a week ago for not showing research.

• Where can I see all my comments on issues on GitHub? — This question is 6 years old and has the third highest number of upvotes on Web Apps SE. It was closed 6 months ago for not showing research.

The issues stated in these questions are clear and understandable to anyone who uses Facebook or GitHub, respectively, and do not require further research to understand or answer them.

If we implement a "research required" policy, we might end up closing most of the questions on this site.

We should encourage more questions and answers instead of imposing more restrictions

A Stack Exchange site with very low activity and user participation does not need stricter question requirements. That would be counterproductive. More restrictions would only reduce the number of questions and answers that are welcome on this site, which would discourage both new and existing users from participating or contributing. This could lead to a downward spiral of declining activity and quality.

I propose that we focus on encouraging more questions and answers instead of imposing more restrictions. This way, we can foster a more active and lively community that can attract and retain more users and, ultimately, enhance the quality and relevance of the site.

Answer 3

To quote two of my favourite Franchises "All this has happened before, all this will happen again" and "Only a sith (or a developer) deals with absolutes"

The healthiest situation is between the two extremes. I also feel like there's also some other 'structural' issues at play here.

Quality answers need quality questions. In the early days there were a lot more low hanging fruit, and standards have evolved over time, and asking the first thing that comes to mind worked.

On the other hand, while I've been guilty of some deeply researched questions in the past (Its a fun rabbithole!) the point of showing your work is so others know where you've been, and what you've tried.

A good question is about clearly articulating your problem - in the old days, before I was a moderator, I used to post a comment going "We need more information to help others help you..."

I also feel like people overcompensate/maliciously comply with showing research.

I think a good starting point is

You should only ask practical, answerable questions based on actual problems that you face. Chatty, open-ended questions diminish the usefulness of our site and push other questions off the front page.

and

inspire answers that explain “why” and “how”

I am going to gank Laurel's example - "What do the red dashed lines mean?"

While I wouldn't give a long story of why I need to know this, knowing what practical problem this solves would be useful.

"I was looking at the island of Aneityum on google maps and noticed a red dashed line. This doesn't reflect any physical boundaries as per the satellite or map images, and it dosen't correlate to any physical boundaries. Google maps dosent' provide an obvious legend - what does this reflect?"

Would be a better start. If OP was going along the coast, and this reflected an intertidal region, it would probably be a matter of life and death.

You don't need a step by step guide to EVERYTHING. It just needs folks to show a little effort.

Answer 4

What level of research needs to be demonstrated for a question to be acceptable?

Let's start with anything at all. Something like:

• I couldn't find the answer in <webapps>'s documentation.
• <webapps>'s tutorial did not cover this.
• I Googled this question, but no clear answer was returned / but the returned answers contradict each other.

Preferably with some examples or links to what they found.
Basically, a question should at least require the bare minimum of effort.

Furthermore, should it become a requirement for users to show some level of research, like searching the help center of the web application in question before posting it? How best should that be demonstrated, and to what degree, if any, must it be established?

Without showing that bare minimum of effort, you just have a question that shows no research. So yes, if we're going to require research, it has to be included in the question.

---

Now, why do I think some research should be shown?

First and foremost, it helps provide context about the user's question. It may just include that one detail that narrows down the question, or it may expose a flaw in the user's reasoning, which resulted in the question in the first place.

A question can't really have too much detail, but even if irrelevant information is added, that could be edited out. You can't edit information into the question when the OP hasn't provided said information.

Learning how to figure out a solution to a problem you're running into is a vital skill everyone needs. Requiring users to do some research might just help them figure out the solution by themselves.

In the end, Stack Exchange sites are not meant to be an all-encompassing database of answers to everything ever. If an answer exists in official documentation, it doesn't need to be on SE.

Answer 5

To the question of "Does Web Apps Require Research?"

I'd say - fine to say it requires research but then how are posts dealt with that don't show research, is the issue.. e.g. suppose an answer was posted, but somebody comes along 5 years later and says no research, so close the question(which then can get deleted)! I don't think that's fair. Or next worst thing is question gets voted to close with no comment as to why. Or gets closed but the comment doesn't say no research, but says "not enough detail". The person that posted it should get a clear notification.. so they don't have a situation where their question gets closed and it completely blindsides them.. While they get stupid notifications any time somebody gives them +10 points but they get no notification when their post gets closed or voted to close! So it's about how policies are implemented.

imposing research requirements might discourage users from asking questions.

I think more of an issue is the treatment of posts of users that have much experience on stackexchange generally, including SO, and the closing of their questions, which particularly happens on webapps, even if they have an accepted answer. Closing can precede deletion, cos eventually a mod can sweep through closed questions and delete them. and then content with some good info can get deleted.

If a new user faces comments asking them about their workings. and gets mad, then stackexchange isn't for them. When I was a new user, I was fine with addressing any comments participating in a question I post, and still am.

What level of research needs to be demonstrated for a question to be acceptable?

Can you show cases where this was an issue? (That will make it much easier to answer that question). It is difficult to come up with a blanket general rule of what constitutes enough research, that covers -any- question.

If there is a dispute over whether a question shows enough research, then the questioner can easily comply with the requirements of the commenter that asks them something.

I've rarely seen a request for some research to be something that'd take a week for the questioner to do!

My issue is, how do you treat a question that is thought to not show research. And old question vs new question.

I've had a question that was voted to close for lack of research, i got no notification of it, there was no comment asking me to add any research. I somehow managed to notice an email somewhere, on whatever email address i've got associated with stackexchange, mentioning that my question had been closed or something like that. No comment, nothing in my notifications.

It really shouldn't be allowed to close or vote to close, without a comment. Some might debate whether a downvote requires a comment, but a vote to close is another level. Especially when it's not simply spam. It's a complete lack of regard to not comment. Infact, really, there should be a comment, and some time for them to reply, before voting to close. The fact users that posts quality questions can have many a number of them closed in such a hidden way is terrible.

can you please state your stance? Should research be required? (yes/no. If yes, how much research?)

I wouldn't be keen to say yes or no 'cos it's so unclear what it'd mean.

And it's such a general question like asking "How do you begin a social interaction" .I wouldn't want to be too formulaic 'cos it'd be unnecessarily restrictive.

What matters is the quality of the question.. that it shouldn't be low quality in terms of "research".

An example of a low quality question in terms of research, would be a question where the answer to it is in the first few Google results and it's clear that the person hasn't made any effort. In which case, a comment can point them to that and the question could be closed.

Trying to force people to state their research can sometimes be a formula for making a question bloated and potentially not as good.. Look at this question I asked on stackoverflow. It's popular. Nobody wants to close it. https://stackoverflow.com/questions/9939760/how-do-i-convert-an-integer-to-binary-in-javascript

Somebody could say "no research" (simply because the question doesn't say anything like "I tried XYZ and it didn't work").

Rather like somebody said on this question of mine where it was voted to close Search YouTube favorites?

But if you look at the stackoverflow question I gave as an interesting case study, many that know javascript could look at that and say "ah that's a good question". Sometimes it can be implicit in the question that somebody has done research. The answer isn't obvious.. The "obvious" thing is to say .toString(2) but that that doesn't actually answer the question since that doesn't cover negative numbers. I could have bloated the question a bit but there's a beauty also in a question not being bloated. My comments on my question, addressing things people say, are not the comments of a lazy clueless guy that hasn't researched anything. And the question is very useful to people.

In contrast, if somebody asked a question that anybody familiar with that knowledge domain knows is an obvious easy question that google would answer straight away, then the questioner clearly hasn't done any research. And the question isn't really useful to people. Somebody should comment on it giving the user some direction and the question should be closed.

I think that's a much better approach, than to say questions should or should not require research 'cos being dogmatic one way or the other is very problematic. Better to understand that some questions don't show signs of a lack of research even if they don't explicitly state much in the way of research, like sometimes a person can ask a good question that possibly a lot of thought went into it and doesn't have an easy obvious answer. And some questions are lazy questions with easily googlable answers. And one shouldn't bundle those two types of question into "no research"! And if it is thought that a question lacks research then it should be dealt with in comment at the time. If the person that asked the question doesn't comply, and the question is just poor quality then sure close it. But don't go retroactively vote to close an old question (and without comment), because it doesn't meet some formula. And that's the danger of a simplistic contentious approach. Things are clearer on a case by case basis. And then we see "ah this is the easily googlable type question where they clearly made no effort". I'd still comment there though both to not be presumptuous(dismissing what might actually be a good question), and to guide the questioner.

Answer 6

Level of Research Required in Web Apps

Due to the large diversity of Web applications and the uses that a single app might have, it depends on the problem the user faces and their circumstances. I.E., google-sheets, the Web Apps SE top tag, corresponds to an online spreadsheet app with features that vary according to the type of user account used; Google consumer accounts have access to certain features, while accounts from specific Google Workspace editions might access to additional features. Google Sheets can be extended and integrated with other systems by end-users in diverse settings, personal, school, small business, government, and corporate.

By research, we understand the time spent by the user on solving the problem faced. Someone with no experience with the problem might have to spend more time than someone with experience. Someone with a high performance in using computers, the Internet and problem-solving might quickly identify what they need to research. Still, someone without these attributes might require a lot of research.

Not all research done by the user should be included in the question. The user should carefully choose what to include to communicate the problem and what was tried effectively and in a way that makes the question attractive to make people interested in answering effectively.

Common Asker Pitfalls

Most of the traffic to Web Apps comes through Google. A Web Apps question was likely shown in the search results, and somehow, the user decided to post a question. Some of these users make pitfalls that I have seen occur repeatedly.

Ignore the Ask Question page guidance

Despite what is said in the modal dialog and the Ask Question page guidance, some users post a plea for help rather than a "practical, detailed question" as referred to in the Web Apps tour.

Let's use the most recent question as an example

How to Detach a Video From a YouTube Playlist

I use YouTube playlists to bookmark videos so I can return to them later. For example, I might have a "Science" playlist for science videos. This lets me find the videos easily if I'm discussing a related topic later with someone. However, when I open a video from a playlist, the playlist remains present on the page. I wish the playlist to go away so I could easily copy the link to that particular video, not the entire playlist. Is there a "detach" button or menu option somewhere that would let me easily detach a video?

As you can see, the OP hasn't mentioned any research done.

Use useless phrases to claim research

Some users think that including phrases like "I have search google and find nothing" is enough to show "research effort". That is just chit-chat that is not providing any value. Something that might be helpful will be briefly describing how the user tried to use Google, DuckDuckGo, or any other web search engine, including a few examples of search queries and a few examples of results explaining why they don't meet the OP's needs.

Ignoring the web app onboarding, getting started, resources for beginners, user interface options and user help.

All the Web apps pointed to massive user base like Google, Facebook and X (formerly Twitter) has simple user interfaces. Some of them have onboarding tutorials to show the users how to start using the basic features and learn about the other features. They all have options or links to access the help center and resources for end users. They should not be ignored as they help the user learn the terms used in the web app, like tweet in X, which is essential to research effectively.

There is no sense in repeating what is on the Web app user interfaces, help center and end-user resources.

Using Web apps as a learning-in-public online notebook

Learning in public is an individual approach to using the Internet to share their development process, hoping to get valuable contributions from friends and strangers. In general terms, it implies making public posts of thoughts and improving them as the thought is being enriched by relations with other thoughts and, in some cases, with research notes.

The problem with using this approach to post a new question in Web Apps is that the new question does not include any research. If they aren't updated they might eventually be automatically deleted if they don't get upvotes / answers or if the people with the vote-to-delete privilege think that the post is not worthy of being kept.

Showing have searched the Help Center

It depends on the problem the user faces and the user's expertise. Most questions asking "Is it possible x?" should be reworded to "How to do x" and include what was found in the help center and why it doesn't meet the user's needs. If nothing was found, the user should describe how the research was done, i.e., mention the help center search and a few examples of the search queries used.

Questions about problems related to the expert use of a tool don't require showing that OP has searched the help center if they are experts on the Web app. This is shown by what is being asked and how it is asked. However, the asker should be aware that the question review could be done by moderators who are not experts on the web app. So, it might be wise to include a brief explanation of the question's complexity or how, being an expert, the OP is still stuck. In most cases, describing what was tried is enough to clarify this.

Answer 7

Research efforts

There are a few interpretations of the question. Some see it as a question about research being reflected in the question, and some as the question proving/detailing specific research that was performed.

I struggle to see this as a binary question. As unpopular as it may be to have a nuanced position, balancing multiple factors seems like the right approach. If referring to the research is important to question then it should be included.

• Yes
  A question should reflect any research necessary to make it a good fit for the site.
• No
  We should not be forcing users to include any specific research simply to “check a box”, that is pointless, and benefits no one.

Insufficient research often leads to closures

I believe any question closure that could have been avoided had the OP performed sufficient research, is also then a closure for a lack of research, regardless of the close reason specified.

Given all the ways we close questions that are tied to a lack of sufficient research, I’m not sure it is even meaningful to argue whether sufficient research should be a requirement.

For example, posting a duplicate question reflects a closure for insufficient research. Posting about something other than a web app reflects a closure for insufficient research. In some cases a question closed due to a lack of detail or clarity may in fact simply reflect a lack of sufficient research, especially if one sees research to include collecting information.

Encouraging Participation

I acknowledge that all quality standards, whether about acceptable user-behavior, content relevance, preferred language, or sufficient research, are necessarily entry barriers to new and continuing participation. These entry barriers are by design and my ideal Web Apps SE encourages participation while retaining minimal barriers to low-quality content.

I would love to increase participation on the site and striving to exhibit a culture of kindness and respect may be more encouraging to user participation than removing or adding any standard.