This post aims to argue for the following belief:

That imposing an evidence or experience requirement is harmful (epistemic status: uncertain)

a) An evidence or experience requirement will not improve answers very much

Just because you had one experience where you did one particular thing and it worked well, it doesn't follow that the technique works in general. Further, often people will judge you without commenting or doing anything overt that you can definitively link back to your action.

If we require academic resources then this would significantly reduce participation due to the time that it takes to find a paper vs. a website, plus paywalling. Also, there are likely to be large gaps in the literature - I doubt that anyone has done any research on whether it is rude to put your feet on the table. Even if there is a paper, you'd need to find/get access to the one, single paper that discusses this issue.

If we require casual resources, ie. someone's advice column, then it isn't actually clear the just because this person managed to get a job at a newspaper that they can give good advice, as opposed to entertaining/controversial advice. It really isn't clear that they are always experts who know better than the rest of us.

There is a lot of value in having multiple answers, so we don't want to discourage them too much. Most people won't be faced with the exact same situation - instead merely one that is similar. Multiple answers allows you to better adapt to the particulars of whatever circumstance you face. This is less likely when we add burdens.

b) The evidence or experience requirement will place a high burden on individuals given:

Social skills is mostly about principles that are "obvious" or "obvious in retrospect", so participants will waste time finding reference to things that everyone already knows.

The best answers will engage in a host of considerations - it will be much easier to be lazy and only engage in one consideration if you need to reference each additional consideration, so we'll receive worse answers.

c) Therefore we should not support this requirement.

The alternative is that we allow people to utilise common sense, to say things like, "It is generally considered rude to put your feet on the table" without having to look for a pointless reference or needing to have an experience where you did put your feet on the table.

Responding to Catja, the alternative to answers backed by evidence or experience is not generalized, generic answers. Instead it is answers that break down a particular perspective on an issue clearly, concisely and precisely, possibly using fictionalised examples as analogies.

  • 3
    So what are you saying -- that people should just offer their unsupported personal opinions in answers and that should be ok? Commented Jul 25, 2017 at 16:33
  • 4
    Can you explain what your exact question is here?
    – HDE 226868
    Commented Jul 25, 2017 at 20:50
  • @MonicaCellio I'm saying that it is completely reasonable to see things like, "It is generally considered rude to put your feet on the table" without a reference
    – Casebash
    Commented Jul 25, 2017 at 23:42
  • @HDE226868: This is meta - it doesn't have to be a question. You can argue for a policy and the answers are people agreeing or disagreeing.
    – Casebash
    Commented Jul 25, 2017 at 23:44
  • @Casebash If this is over our little altercation, I never said that you need a reference. I just said that it needs to be supported by that or experience, and that you could improve your answer in that way.
    – Zizouz212
    Commented Jul 25, 2017 at 23:48
  • @Zizouz212: Well, I don't have an experience of putting my feet on the table either
    – Casebash
    Commented Jul 25, 2017 at 23:49
  • 3
    Possible duplicate of Why doesn't this site have a back it up rule?
    – user288
    Commented Jul 25, 2017 at 23:53
  • 1
    @Casebash I don't want to discuss or argue this in the comments. You have two answers presented to you in response to your post.
    – Zizouz212
    Commented Jul 25, 2017 at 23:53
  • @Hamlet I personally disagree that this is a duplicate. They're different questions entirely, and have attracted different responses.
    – Zizouz212
    Commented Jul 25, 2017 at 23:55
  • @Zizouz212 I was merely responding to your comment. I don't think you have anything to complain about. I'll add my thoughts on the other two answers too - you aren't forced to respond if you don't want to.
    – Casebash
    Commented Jul 26, 2017 at 0:09
  • 2
    @Casebash It doesn't have to be a question, per se, but it should have a focus - and I'm not sure what kind of discussion it's trying to encourage. It reads like a person's thoughts without asking for any sort of response.
    – HDE 226868
    Commented Jul 26, 2017 at 0:45
  • You make a lot of claims without referring to any specific examples of answers you feel are acceptable. Until we see examples it is unlikely we're thinking about the same thing.
    – Catija
    Commented Jul 26, 2017 at 0:46
  • @HDE226868 The title say "Against the evidence or experience requirement" - I think it is pretty clear.
    – Casebash
    Commented Jul 26, 2017 at 0:52
  • 4
    I'm voting to close this question as off-topic because there isn't a question or attempt to start a discussion. This content would be better under one of a number of relevant meta discussions.
    – HDE 226868
    Commented Jul 26, 2017 at 1:03
  • 2
    @HDE226868 Not all meta questions have to have a "question" in it. It can also have suggestions or announcements put forward, to which the community can raise their concerns or leave feedback in the answer boxes. It's tagged 'discussion'. Let's discuss, not close it. So far, I think this post is doing well. We got 2 good answers to it. This question should stay.
    – NVZ
    Commented Jul 26, 2017 at 8:53

2 Answers 2


We are a site that is largely subjective. This, on its own, goes counter to the SE model requiring objective content. As such, we must strive to make our subjective content as objective as we can by holding these questions and answers to an extremely high standard. Everyone has an opinion for how to fix a certain interpersonal situation. We don't want opinions. We are not a discussion board/forum/help site.

So, to help in that, we should rely on the guidance that Stack Exchange gives us regarding subjective content and on the users (particularly the moderators) and policies of sites that have already gone through this set of issues like The Workplace and Parenting.

How does SE guide us?

Firstly, we should look to the (old but still valid) Stack Exchange Blog post "Good Subjective, Bad Subjective".

Thus, questions that are not answerable — discussions, debates, opinions — should be closed as subjective. It seems simple enough: Fact good; opinion and discussion bad. But why?

Most forums and chat rooms have a scale problem. As in, they don’t. The more people that join the discussion, the more noise each of those connections bring. So the forums get progressively noisier and noisier, and suddenly one day … you stop learning.

This shows us the heart of the issue. When everyone is just posting what they think someone should do in a situation, we end up with dozens of less-than-helpful answers that aren't much use to the person who asked the question. Remember, the users of SE expect good content that's supported with evidence. Giving them anything less than that breaks the social contract of SE and leaves this site open to removal by the Powers That Be.

The folks at Moms4mom owned up to the subjective issue and came up with a set of principles to create useful subjective discussions on parenting: the Back It Up! Principle. Back It Up! means that your answers must be based on either:

  • Something that happened to you personally
  • Something you can back up with a reference

They talk about how “opinion, by itself, is noise.” They’re not saying that subjective opinions are to be avoided; they’re attempting to mold and shape their inherently subjective Q&A; into something constructive, informative and helpful.

Please take careful note of this Post and read the rest of it. I don't want to reproduce the entire thing here but we need to follow this guidance. If we do not, the site will likely be closed down. There haven't been any sites closed recently (that I know of) for subjective content but a few sites have been at risk of it or were outright closed in the past. We can not hope to keep this site around if we allow unfounded content to rule here. I will duplicate the six bullet points regarding how to write a good subjective question and what should be expected in an answer:

  • Great subjective questions inspire answers that explain “why” and “how”.
  • Great subjective questions tend to have long, not short, answers.
  • Great subjective questions have a constructive, fair, and impartial tone.
  • Great subjective questions invite sharing experiences over opinions.
  • Great subjective questions insist that opinion be backed up with facts and references.
  • Great subjective questions are more than just mindless social fun.

Just this morning I received an email from SE about this beta (It was sent to users who appear on the leaderboard). It talks about how it's time for us all to focus on the rough edges of a site and shaping it into the site we want it to be:

By now you've probably noticed some rough edges to the site: questions that don't seem to fit, tags that are lacking or ambiguous, answers that launch tangential discussions... Well, Stack Exchange sites are defined and run by their users, and we've put together some tools to help you discuss and refine your site:

While "The 7 Essential Meta Questions of Every Beta" is a bit out of date, we're working on the important questions mentioned there. The other tools are this site (IPS Meta) and Chat - where you've already mentioned your concerns about this. The last part of the email gives the directive:

  • Vote up useful posts, and vote down poorly researched ones.
  • Edit questions and answers that are badly worded or formatted to improve them.
  • Close questions that are unclear or off-topic.
  • Flag anything that's spam, off-topic, contentious, or just needs special attention... one of the moderators will check into it right away.

How do our sister sites help?

Our fellow subjective content sites are a great place to glean advice on what to do and (occasionally) what not to do. The posts on their metas are a resource to us. Here's a question on Workplace Meta from when they were in the same position we are now:

Do we have a Quality control issue?
We have a few questions that while not off topic I do not think are high quality questions as asked but have been upvoted:

These questions are broad and generic enough that they hit alot of peoples buttons. But I think they are too general to be good questions. They get up votes because people can relate not because they are good questions.

We have great participation and the one question that was clearly a rant as a question was closed off quickly. But these questions being voted up during private beta scream "DANGER WILL ROBINSON" to me.

Community Manager Robert Cartaino answered it.

My biggest concern for this site is that these overly-broad "how can I be more awesome?" questions will become the mainstay of this site. It's not that they're inherently off-topic — but you don't stand a chance at making the Internet a better place by asking generalized, generic questions that have been asked 100's of times on every other site on the topic… and the trite, hackneyed answers they attract aren't going to be all that interesting, either. [ . . . ]

The remediation is to stay vigilant. Guard jealously the core purpose of this site. If a question can be improved, improve it. If you have concerns, leave a comment or start a meta post. If a question is not a good fit, close it.

But don't be rude or overly curt. You can be a bit more strict early in the beta, because these earliest days are more about setting the right path than getting everyone's questions answered. But take every opportunity you have to discuss why you took an action or feel a post or behavior might need consideration. Leave lots of signposts to help guide users who might not "get" what this site is about.

What does it all mean?

So, I think we need to do two things and we need to do them extremely well, particularly right now.

  1. We must require questions to be about extremely narrow, specific issues.
    These issues should be related directly to a type of relationship, country and be centered around a specific example interaction that needs to be addressed. If they do not, the questions must be closed until they do.

  2. We must require answers that address the specific question asked directly and these answers must be supported with either personal experience or a reference.
    Answers that do not meet this requirement should be either downvoted, commented on for improvement, or outright deleted. Once we have moderators, they can use post notices to request improvement.

Failure, particularly at the outset, to protect this site and the core of what is on topic will put us at risk for being extremely low-quality and possibly removed entirely. We are fighting for survival.

  • 1
    Note that this answer isn't intended to explain how to write a good answer to a subjective question. For information on that, see Hamlet's related answer here.
    – Catija
    Commented Jul 25, 2017 at 19:42
  • 2
    +1 Thank you. giphy.com/gifs/1Z02vuppxP1Pa/html5
    – NVZ
    Commented Jul 25, 2017 at 19:58
  • @NVZ Haha that was classic!
    – Zizouz212
    Commented Jul 25, 2017 at 20:02
  • Thanks, for the comprehensive answer. "By now you've probably noticed some rough edges to the site: questions that don't seem to fit, tags that are lacking or ambiguous, answers that launch tangential discussions..." - this is just a default email scheduled to be sent at a specific time in the lifespan of a site... It does have a sentence mentioning "poorly researched posts" - but what are these? Is a post without a reference "poorly researched" or is there a common sense exception?
    – Casebash
    Commented Jul 26, 2017 at 0:11
  • "You don't stand a chance at making the Internet a better place by asking generalized, generic questions that have been asked 100's of times on every other site on the topic" - I think that we've already done a good job of excluding overly broad questions so that these don't dominate the discussion. So I can't see how point 1 is directly relevant to this discussion. Maybe it is, but the link isn't clear.
    – Casebash
    Commented Jul 26, 2017 at 0:13
  • "Failure, particularly at the outset, to protect this site and the core of what is on topic will put us at risk for being extremely low-quality and possibly removed entirely. We are fighting for survival" - I disagree with this characterisation. I believe that the site is doing quite well at the moment and it is already far more interesting than any other similar sites I've seen on the internet because it focuses on specific answerable questions so that you don't have the same generic questions being asked over and over again. On the contrary, intervention seems risky...
    – Casebash
    Commented Jul 26, 2017 at 0:18
  • 4
    @Casebash curating good questions is only half the battle. You can't allow generalized, generic answers, either.
    – Catija
    Commented Jul 26, 2017 at 0:19
  • @Catija I don't think that is an accurate characterisation of the alternative. The alternative are answers that break down a particular perspective on an issue clearly, concisely and precisely, possibly using fictionalised examples as analogies.
    – Casebash
    Commented Jul 26, 2017 at 0:25
  • 1
    @Casebash The underlying motivation behind most of this is that we don't want screw ups happening again. All we're really asking is if you have the experience to answer the question. Experience here should be interpreted broadly here - you need to have some familiarity and experience with the issue, or the culture, or the society... Pure hypotheticals such as "If I were in this situation, I would do this..." have been empirically proven to be detrimental to the site. Such answers aren't bad, as long as they provide their arguments, and reasoning behind them, as I outlined in my answer.
    – Zizouz212
    Commented Jul 26, 2017 at 0:32
  • 3
    @Casebash you would have an easier time arguing your point if you posted examples of answers that meet your description. Right now we're tilting at windmills.
    – Catija
    Commented Jul 26, 2017 at 0:33
  • 5
    @Casebash You make broad, unsupported statements and pass them off as facts. Most of them aren't really worthy of response... requiring support does not make people lazy... people who write unsupported answers are already lazy.
    – Catija
    Commented Jul 26, 2017 at 1:26
  • 4
    @Casebash You don't think I should be pointing out that your argument is lacking in details? I've pointed this out several times and yet you have not edited your question to clarify, to cite examples, to source your assertions. You expect a mod to set rules for the site (which they do not) but by the same token you don't accept someone who might be a mod pointing out where you need to improve?
    – Catija
    Commented Jul 26, 2017 at 2:10
  • 5
    @Casebash I am open minded. I'm asking you to prove that your assertions are true whereas you're asking me to accept them simply because you say they are true. If you can prove that what you say is corect with evidence, I will happily accept it but every source I've found on all of the subjective-ish sites on SE have come to the conclusion outlined in this answer. I am a scientist at heart, so I love having my beliefs tested. Please, challenge me!
    – Catija
    Commented Jul 26, 2017 at 2:23
  • 5
    @Casebash You've presented no evidence, just hypothetical statements on your part, so there's no evidence to consider. You've barely addressed or responded to any of the points mentioned in this answer - only propelled your thoughts and beliefs and wondering why we don't accept/consider them.
    – Zizouz212
    Commented Jul 26, 2017 at 2:29
  • 1
    @Zizouz212 I don't expect you to accept them, I expect you to consider them.
    – Casebash
    Commented Jul 26, 2017 at 2:32

I want to answer two simple questions here:

  1. Why do we need an evidence requirement?
  2. How easy it is to follow it.

Why do we need an evidence requirement?

Like Catija's answer, Stack Exchange communities are built on the premise that questions will be answered that are strengthened by evidence. Since we're a subjective site (not something such as Law, or Mathematics that both have referenced, objective answers), our "evidence" will be different. Answers need to be backed up by something that happened to you personally, or by an authoritative reference. The latter won't always be possible, so we should concentrate on the former.

One of the reasons why I'm very invested in this is because of cases like these (barring AJs and NVZs answers). Where many of the answers provided that are just what people think - and are very clearly inappropriate. As for why these answers got up votes, I tried to explain that phenomenon in another post.

A requirement places the onus on the answerer to prove to the community that their answer is relevant and appropriate for the situation described in the question. Overall, it would support higher quality answers within the community.

How easy would it be to follow?

Very easy.

For answerers:

Hamlet and I have been talking a fair bit about what to do with this policy and make it work for everyone. If you want to take a look at our conversation, feel free to find the transcript in The Awkward Silence. I'd like to credit him for most of this.

Answers have two parts: their argument, or the advice that you are trying to provide to the asker, as well as a support, which is a statement saying why your argument is applicable to the situation.

An answer that includes both is a good-quality answer (note how I am saying good quality, and not necessarily appropriate). Only including one or the other doesn't make for a great answer. How do you structure your support? I'm going to quote Hamlet in our conversation (with some terms changed):

Here's an example of an argument. "I was in an abusive relationship. Nothing I did improved things. I later found out from a psychologist that this is normal. Meeting with the psychologist helped give me courage to walk away. I suggest that you meet with a psychologist, or someone who is a professional who can offer advice. Extra sentences."

It's a good argument. What's it missing? Support!

Specifically, how do I know by reading this whether I'm also in an abusive relationship? I would know by the information that the writer gave about their situation, and I could compare that information to my own experiences. Here's an example of something that isn't an argument.

"I was married for fifteen years, and my spouse and I started having arguments about what to watch on TV. To solve this, I divorced my spouse."

It has support. I can tell by reading it whether it's applicable to me.

But no argument. Nothing to convince me that if I were in the same situation, that divorcing my spouse would be the right thing to do.

Not every situation is the same. As I wrote here, something that worked for you may be completely inappropriate for the asker. By answering, you're trying to make the internet a better place. We want to make sure that anyone who has a question can come here, and get the answer that is most appropriate for them.

  • +1 for this, but what's wrong with the linked question? My answer provides an experience-based advice there. Isn't that the sort of answers we require? Or am I missing something?
    – NVZ
    Commented Jul 26, 2017 at 8:50
  • @NVZ Oh oops. I was more referring to the initial three or four answers that it got (almost everything except yours and AJs), that were all a disaster. I kind of forgot to say that - I'll change it right now :)
    – Zizouz212
    Commented Jul 26, 2017 at 18:15

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