Back when an issue with the tag came up, Monica Cellio made a suggestion:

Re the addendum, would [hard-science] better convey what you're after?

The rationale behind the question that started that discussion was that almost all answers are expected to be based in science and logic.

Well, the tag was started, I asked a question to test it out. Serban Tanasa's answer earned a bounty I placed, but I didn't accept it, because it wasn't as in-depth as I wanted. I also posted an answer to demonstrate just how I thought answer using the tag should be.

I hope nobody takes this the wrong way, but I don't think that the answers have turned out to be quite as good as I had hoped. Looking at the tag wiki excerpt,

All answers to this question should be backed up by equations, empirical evidence, scientific papers, other citations, etc.

To this day, I have yet to see more than a few answers - and this takes into account that some of my answers aren't up to snuff here - that actually fit these criteria.

In short, I see the tag going the way of , which is not good. I tried to bring this up in Defining the boundaries of the "hard-science" tag, but the issue has not gone away. I see questions migrated away to other sites, like Physics and Engineering. That's good, but I think we should strive to have questions demanding tough answers on-topic and welcome here.

Perhaps I'm just getting a bit touchy about the issue because of the types of topics I'm used to writing about on Stack Exchange. But I think that this is a necessary issue. After this, I'm not going to push the point. I promise.

Can we please, please bring back to the way we envisioned it, and make sure that the original ideas - like what an asker should want in order to use the tag - are enforced? If things keep sliding, the concept, I think, will be dead.

  • $\begingroup$ Related: worldbuilding.stackexchange.com/questions/19000/…. That serves as a base for the types of answers I'm looking for here. $\endgroup$
    – HDE 226868 Mod
    Commented Jun 11, 2015 at 21:44
  • $\begingroup$ Also worldbuilding.stackexchange.com/questions/19044/… $\endgroup$
    – HDE 226868 Mod
    Commented Jun 12, 2015 at 16:03
  • $\begingroup$ @James hard-science only defines the rules for answering a question, not the plausibilty of the questions topic. If the topic is in opposition to science then the answer scientifically explains why it is wrong. $\endgroup$
    – anon
    Commented Nov 20, 2018 at 22:03
  • $\begingroup$ The underlying problem here is that "hard science" is a real phrase that exists in the real world, and it rarely (or never?) means what the "hard-science" tag means by it. So we're fighting an uphill battle in convincing everyone to understand this tag in the way it wants to be understood. If you want to have a tag that means what the "hard-science" tag means, I think the best approach is to rename it to something that doesn't mean something completely different. $\endgroup$
    – ruakh
    Commented Jan 1, 2019 at 0:06
  • $\begingroup$ (As proof, I point to the highly-upvoted answer below that uses the phrase "hard-science SE sites"; and your own comment to it, which does the same. There are no SE sites that require all answers to "be backed up by equations, empirical evidence, scientific papers, other citations, etc."; rather, the phrase means SE sites that specialize in so-called "hard sciences", such as physics, chemistry, and biology.) $\endgroup$
    – ruakh
    Commented Jan 1, 2019 at 0:08

3 Answers 3


I think, to prevent going the way of , we're going to have to be strict about things. Really, quite harshly strict about it - like this:

  • If a question is not a question but tagged as one, a tag edit is needed. Any and all attempts to re-add the tag should be rolled back.
  • If an answer to a hard-science question doesn't cite sources, have evidence and proofs, then (this is the harsh bit) it is not an answer and thus it should be deleted.
  • (perhaps this might go a little too far) If an answer is scientifically wrong, it is also not an answer and should be deleted. Hard science expects hard, verifiable, correct science.

I note, as per HDE's comment, that bullet point 3 is outlawed on many of the hard-science SE sites - so I'm open to that bullet being rejected here. Just a suggestion.

Now, in my personal opinion, does serve a useful purpose here, for those questions that really want completely scientific answers to problems that might be off-topic on a hard science SE site. If it goes the way of , it will no longer serve a purpose. Thus, I think we should apply the measures detailed above, and apply them consistently and strictly.

I expect some people to disagree. That's fine, but this is an issue we need to solve, so if you wouldn't be OK with helping apply these policies, let's debate them. Post a counter-answer here or a comment, and let's actually sort this one out.

  • $\begingroup$ Regarding bullet point 1: I've considered doing this, but I've always kind of felt that the use of the tag is at the discretion of the asker, because it defines the sort of answers they want. The tag influences the answers more than the question, though if the question goes against science, then the tag serves no purpose. I disagree with bullet point 3; that practice has been controversial and generally outlawed on hard science sites. Just because something is wrong doesn't mean that it shouldn't be an answer. So I actually strongly disagree with that bullet, and possibly bullet point 2. $\endgroup$
    – HDE 226868 Mod
    Commented Jun 11, 2015 at 21:52
  • 1
    $\begingroup$ @HDE226868 I agree on point 1, the idea was more if there's a question that is obviously against the spirit of the tag, it should be edited out. I'd be giving a very wide latitude of discretion to askers here. $\endgroup$
    – ArtOfCode
    Commented Jun 11, 2015 at 21:58
  • $\begingroup$ In those cases, I totally agree. $\endgroup$
    – HDE 226868 Mod
    Commented Jun 11, 2015 at 21:58
  • $\begingroup$ I totally agree with point 1 partially with point 2 and maybe with point 3 The point 2 goes too far in my opinion. If the content of the answer is good enough, it should not be deleted even if it lack explanations/citations/proof. It's like the short anwsers we get on some questions. They are not bad but usually don't deserve an upvote either. If we feel the person was just very lazy maybe it could be downvoted for low quality. The problem in my reasoning is that a majority of users might upvote it because it's a good idea or because it's funny :( $\endgroup$
    – Vincent
    Commented Jun 11, 2015 at 22:46
  • $\begingroup$ Point 3 : comment on the issue and downvote, hoping the user will correct the answer or delete it by himself. This point is not specific with the hard-science tag (but maybe it's not a problem for some tags like magic). Having informations that are false is not acceptable even for Worldbuilder SE, for the exact same reasons as for the other SE. We are supposed to have quality questions and answers that are meant to be used by future users. We can't allow things like 1+1=3 (unless you can prove) or saying that expanding the ocean as an effect on meridians (?). $\endgroup$
    – Vincent
    Commented Jun 11, 2015 at 22:57
  • $\begingroup$ Deleting this content should be considered seriously as a last mean. $\endgroup$
    – Vincent
    Commented Jun 11, 2015 at 22:58
  • $\begingroup$ Point 2: @Vincent I say delete because if it doesn't include all that, it doesn't answer the question. $\endgroup$
    – ArtOfCode
    Commented Jun 11, 2015 at 23:00
  • $\begingroup$ Technically yes, they do answer but because of the tag stiffness: no. That is your reasoning ? $\endgroup$
    – Vincent
    Commented Jun 11, 2015 at 23:02
  • $\begingroup$ If this is so, maybe your right. As a proposition we could state in the tag description that answers that do not meet the criteria will be downvoted and/or deleted. Can we do that? $\endgroup$
    – Vincent
    Commented Jun 11, 2015 at 23:11
  • $\begingroup$ @Vincent that's my thinking, yes. A note in the wiki is a good idea. $\endgroup$
    – ArtOfCode
    Commented Jun 11, 2015 at 23:18
  • 5
    $\begingroup$ Wrong answers that are (otherwise) still answers should be downvoted, edited if you can fix it, and/or commented upon to guide the author. Merely being wrong is not grounds for moderator deletion and moderators are not arbiters of correctness, though the community is free to delete answers it feels do not belong. $\endgroup$ Commented Jun 12, 2015 at 0:38
  • 2
    $\begingroup$ I am not as experienced as any of you, but as Monica said, deletion should not be the rule. It is probably better to comment, to try to motivate the answerer to improve xir answer. And if all active people were set on downvoting it, that would at least, clear it out of the way. $\endgroup$ Commented Jun 12, 2015 at 7:03
  • 2
    $\begingroup$ I agree with Monica and @bilbo_pingouin. Deletion is almost certainly going too far, and mod-deletion is certainly going too far for simply failing to meet the [hard-science] criteria. However, downvoting such answers is perfectly reasonable, as they do not answer the question within the constraints established by the question. Deletion is more for answers that really don't answer the question, including link-only answers. $\endgroup$
    – user
    Commented Jun 12, 2015 at 7:29
  • 2
    $\begingroup$ I've noticed some sites enforce a disclaimer for some questions, and have special "close-like" stuff for answers. Like, Skeptics SE, IIRC. I'm being vague because I'm not sure I recall this exactly, sorry. Would such a thing help here? (disclaimer: didn't read all the comments) $\endgroup$
    – o0'.
    Commented Jun 21, 2015 at 20:08
  • 1
    $\begingroup$ @Lohoris Nice idea. That would be good. $\endgroup$
    – HDE 226868 Mod
    Commented Jun 21, 2015 at 20:12

I agree with ArtOfCode, but wanted to add that many of us don't feel qualified to answer . I suspect the same would be mostly true with regards towards knowing is something has enough hard science in.

Downvoting answers and adding comments is still the initial first step for answers that are not good enough. The deletion reason for answer would be "low quality". Scientifically wrong answers that still (inaccurately) use "facts" and equations (however misapplied) are in fact answering the question, and as per norm should just be downvoted.

I have a feeling that those of you who constantly use the tag are going to have to spend a lot of time adding comments, VTC, VTD, editing and quoting the tag wiki to get this into the heads of most of our members, since while meta is where decisions are made, it is not always read.


I am bumping this as it now seems the original idea of this tag has been perverted through a misconception of its purpose.

As I understand it, use of this tag is purely to require answers to support all critical points with scientifically backed sources or equations.

I whole heartedly support this notion. Some topics are only worth answering under full scientific scruitany. Such questions should allow the possibility that they are impossible, however all answers must be supported.

What I do not believe this tag ever supported, was that hard-science questions be only hard science topics.

Lets analyze this under the lens of an obviously contested topic: Anatomically correct succubus: the diet.

This question asks if a humanoid organism could sustain itself through intercourse with a human, inkeeping with the mythology of the succubus.

Obviously a succubus is a mythological creature and therefore does not exist in science. But that is not the question.

The question is if it is plausible that a humanoid organism could sustain itself this way? This could easily be explored under a strong scientific lens.

For instance:

One answer could reason that succubus is a hummanoid so it will have similar caloric needs to a human [insert links and equations]. It could then point out that ejaculate only contains 22 calories [insert supporting link] it could then mathematically demonstrate that it would be impossible for an organism to survive on such limited nutrition.

Such answer satisfies the criteria of hard-science and yet can still be wrong. Because the question assumed the only obtainable substance during sex was ejaculate instead of considering other things like blood. Its still a valuable answer because it outlines variables in the system to explain.

You might be thinking why I went with that example:

My point was does not dictate the correctness of the answer or even the question, that is the purpose of comments and voting. Hard-science does not dictate the subject matter of the question or answer. What makes hard-science valuable as a tag is that it requires all presented variables to be scientifically supported so that they can be mathematically evaluated.

Going back to the example, because the answer was forced to backup all its points and provide the equation a better answer could emerge that accounted for factors not previously considered.

Now if anything this does impose some restrictions on questions. This means that questions must accept the possibility they are impossible and allow for answers that explain theyre impossible.


Science is not a process for outright dismissal of claims. Science is the human process by which to convert claim to fact or fiction by answering why or why not based on equally proven facts.

  • 1
    $\begingroup$ "For questions that require answers based on hard science, not magic or pseudo-science, but do not require scientific citations." - if you do not require scientific citations, [tag:science-based is for you. "If a question is not a hard-science question but tagged as one, a tag edit is needed." - from highest-voted answer here. $\endgroup$
    – Mołot
    Commented Nov 22, 2018 at 14:13
  • $\begingroup$ I disagree with you in that I have always maintained that to expect a hard-science answer you need to ask a hard-science question. Thats not what your question was, your question was 'explain how this concept would be implemented.' A hard-science question would be in the form 'prove with citations or equations if/how this specific implementation works.' You have to provide a specific implementation to prove or disprove, which you didn't do. Compare yours to the specificity of worldbuilding.stackexchange.com/questions/22769/… $\endgroup$
    – kingledion
    Commented Nov 29, 2018 at 4:00

You must log in to answer this question.

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