The Problem

I was going to write a question here asking what the appropriate response is to incorrect upvoted answers, but I realized that I'm far from the first person to ask this, e.g.

How should we deal with upvoted incorrect answers?
Are technically wrong answers and comments allowed on physics.SE?
Flags and technical innacuracies
physics.SE's inability to deal with users who are highly persistent...
Does it matter if answers are correct?
old questions with incorrect answers

Many questions on the scientific stack exchanges are highly advanced, and only a small number of people in the world are likely to actually know the answer. However, especially if the answer has the right "popular" flavor to it, it might garner a lot of up votes from people who can't judge it's merit, but know that it sounds good. When I encounter an incorrect answer like such, I've found my options to be very unsatisfactory:

  • Down-voting: The expert to non-expert ratio of voters makes this ineffective.
  • Commenting: Often-times the ways in which an answer is wrong cannot be succinctly summarized in a comment, and often-times comments on older questions get buried below a stack of previous comments, getting no visibility.
  • Answering: Sometimes it's necessary not just to answer the question, but to rebuke the incorrect answer. Having a long rebuke section in an "answer" doesn't seem appropriate. More-over, such a rebuke is unlikely to get the same upvotes as a nice-sounding answer since it likely brings attention to the nuance and complexity of a subject. This seems like an ineffective mechanism against incorrect answers that already have a lot of vote-mass.

In my view this is a problem, and has led me to think that stack exchange is an unreliable and potentially misleading source for these types of advanced subjects (although that hasn't kept me from coming back ;)

So I thought of a potential remedy:

A "Requires Expertise" Notice

For highly specialized questions, the notice would remind users that they should only answer or vote on questions if they have the expertise to judge whether an answer is accurate or not. I'm thinking of something similar in style to the "controversial post" notice. I'd think that this could be added by mods, or perhaps the questioner themselves, who recognized a question was likely to garner such votes. To be clear I'm not suggesting certain users be labeled "experts" or anything like that or that anything here would be reputation based, just a notice in an attempt to dissuade irresponsible voting & answering.

Edit Whether it's a good idea or not, I fear the suggestion is being misunderstood, so I've tried to remove some of the fluff in an attempt to clarify.

  • $\begingroup$ Would any downvoters like to explain why they don't like the idea? $\endgroup$ Commented Jan 30, 2019 at 5:58
  • $\begingroup$ I don't think it's a problem worth trying to find hacks to resolve. $\endgroup$
    – Kyle Kanos
    Commented Jan 30, 2019 at 12:01
  • 2
    $\begingroup$ I don't see how this would help? Either we restrict posting on "expert" questions to people we "know" are experts (likely shutting out real experts who aren't active on SE), or we basically allow people to deem themselves experts, leading to basically the same situation as now where we get hit by the Dunning-Kruger effect. $\endgroup$
    – JMac
    Commented Jan 30, 2019 at 12:15

3 Answers 3


Note that one of the bounty reasons is

Authoritative reference needed

Looking for an answer drawing from credible and/or official sources.

I've always interpreted such bounties as evidence that the bounty-setter believes the existing answers, even if upvoted, are fatally flawed. Offering such a bounty has the dual effect of (a) implying that the existing answers are inadequate, and (b) featuring the question in a place where a subject-matter expert is slightly more likely to happen across it.

  • $\begingroup$ This is good and I didn't know about it. Basically what I'm suggesting is something like this could be put up without sacrificing reputation in making a bounty. $\endgroup$ Commented Jan 30, 2019 at 17:40

Every post on our site requires a certain level expertise to evaluate. Sometimes that expertise is having completed primary school, sometimes it is having a dim recollection of high school physics, sometimes it is having studied a particular subfield of physics in depth. But you cannot understand any post about science if you do not share a basic framework of knowledge with the author.

In an ideal world, users always should only vote on posts if their knowledge of the general topic of the post is such that they can evaluate its merits. In the real world, they don't. But this is not specific to "highly specialized questions", people will vote on questions regarding Newtonian mechanics whether or not they know Newton's laws. All this notice would achieve is give the questions tagged with it an air of elitism (like "Oh, for the other posts, it's fine if you just vote on it, but for this one, you actually need to know something").

Therefore, we shouldn't do this. It is SE's basic philosophy that people vote on what they find useful, and you cannot prescribe to people what they find useful. Some people enjoy reading walls of text, others enjoy reading detailed mathematical derivations, others enjoy reading posts about topics they have not the faintest clue of. That's just the way it goes.

  • $\begingroup$ Most people like pictures though. $\endgroup$
    – Kyle Kanos
    Commented Feb 1, 2019 at 3:17
  • $\begingroup$ It's my honest experience on this site that the problem is specific to (or worse with) more specialized questions. I guess I'm in the minority for thinking that. I figure a proper phrasing could also probably avoid the notice sounding elitist, but it's probably to late to salvage this suggestion. I probably came off as somewhat elitist myself in my original question as well. $\endgroup$ Commented Feb 4, 2019 at 9:45

Notices under questions (popular ones include the 'controversial post' notice on Skeptics SE and the 'resource recommendation' on Physics) are added manually by mods and they need to be removed manually too. This can't just show up for certain users 'when they try to vote', and there's no provision for something with an 'okay box'. This requires changes to the SE code base, and would need to be proposed on Meta SE.

Additionally, this would actually be applicable everywhere, since however rudimentary the question is, votes should only be cast when someone knows that the content is accurate. So adding it as a notice isn't sensible. It's more logical as an addition to the help center, and something similar's already there:

Whenever you encounter a question, answer or comment that you feel is especially useful, vote it up! (From the privileges section)

  • $\begingroup$ I fear the "popup" part, which I viewed as secondary, is being a distraction, so I removed it from the question. I also tried to make it more clear that I was indeed referring to something that would be manually added, and that it would only be relevant for the most specialized questions which the majority of physics stack exchange users aren't equipped to answer or judge (clearly I'm not suggesting it be put everywhere). $\endgroup$ Commented Jan 30, 2019 at 17:45
  • $\begingroup$ @aquirdturtle I know you aren't suggesting it should be put everywhere. I'm suggesting it would be applicable everywhere. Why do you want to remind people not to follow bad practices only when they're looking at more sophisticated posts? $\endgroup$
    – user191954
    Commented Jan 31, 2019 at 0:58
  • $\begingroup$ Because I don't think it's a problem for easy questions, I think it's a problem where "the majority of physics stack exchange users aren't equipped to answer or judge". I don't think you need to remind people where they don't seem to need reminding. $\endgroup$ Commented Jan 31, 2019 at 6:21
  • $\begingroup$ @aquirdturtle When there's an easy question, there's someone who doesn't know the answer to it. The people who vote here aren't just physics SE users: there are people from other sites who just read and vote, and there are people who come by for HNQs. HNQs are easy usually. Voters often need/have no discipline-specific knowledge. Moreover, regarding "majority of physics stack exchange users", that means that it's subjective: who decides if the topic is advanced? Mods? OPs? Who even knows the level of education of a majority of users? $\endgroup$
    – user191954
    Commented Jan 31, 2019 at 7:34
  • $\begingroup$ I agree with everything in your first 4 sentences, of course. But as I said, I don't think there's a similar problem for "easy" questions, so the points seem irrelevant. Of course I agree that this is subjective as well, which is why all I'm suggesting is a warning. And as I already said, it makes sense to me that both mods and OPs could set such a warning. $\endgroup$ Commented Jan 31, 2019 at 7:44
  • 2
    $\begingroup$ @aquirdturtle mods aren't permitted to use their powers to decide factual accuracy, and that's something which is empirically justifiable (through citations etc). They can't decide if a question is 'hard' or not. On some sites (I have one science site in my mind), there are mods who don't even have degrees in the subject, so they may not know how complex a topic is. And obviously OPs can flatter themselves by saying they're asking nontrivial questions. So to me it makes sense that nobody has the 'authority' to set up such a warning, even if you argue that it's not universal. $\endgroup$
    – user191954
    Commented Jan 31, 2019 at 7:48

You must log in to answer this question.

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