A recent question posited a fantastic condition then used the tag. In essense, the author wanted a science-based solution to a problem that wasn't science-based in the first place.

We've done a good job of clarifying how the three science tags word (, , ). Those three tags are almost unique in that, unlike nearly every other tag, they constrain the answer rather than categorizing the question. But I've encountered a number of questions over the last six months where the question failed to meet the expectations of the tag (i.e., failed to include details, specifications, limitations, or conditions that an answer conforming to the tag would require.

That might be reasonable, but my gut suggests that it's not. How, for example, other than out of shear luck, can a question tagged be answered if it doesn't provide the conditions and requirements such an answer would demand?

So my question is this: Should we edit the three science tag wikis to require questions that use the tags to also meet the expectations of the tags.

  • $\begingroup$ You do not include internal-consistency in the tags to edit? $\endgroup$ Commented Jan 4, 2023 at 1:38
  • 1
    $\begingroup$ @Tortliena internal-consistency self-solves the problem. By definition, the OP must provide the relevant rules and a scenario to test those rules with. Thus, I didn't include it. $\endgroup$
    – JBH
    Commented Jan 5, 2023 at 2:23
  • $\begingroup$ Why is this any different than closing a question for needing more details? $\endgroup$
    – Topcode
    Commented Jan 5, 2023 at 2:31
  • $\begingroup$ @Topcode Recognizing that few people take the time to read wikis, this is an effort to codify an expectation. As I mention in my post and Tort mentions in his answer, below, it's possible for people to get lucky and have an Q that's answerable to the expectations of the tags without the Q meeting those same expectations, but most don't. By enhancing the wikis, we have the ability to VTC and point to the wikis for, frankly, a superior rationalization, because sometimes it's not the details that need to be added, but the tag that needs to change. $\endgroup$
    – JBH
    Commented Jan 5, 2023 at 2:49

2 Answers 2



Unlike other tags whose purpose is categorization, these tags set intentions : loosely real-world based solutions (), a better than suspension-of-disbelief need (), and a very accurate world (). Thus, people need to provide the material to reach these intentions; If they don't answerers can miss the target. After all you can't reasonably ask someone to build a sturdy house 🏰 if all you bring them are sticks or worse, straws 🛖!

... But with some conditions

That more clarifications would change answers

There are a few things to consider though. Most importantly, it's to check whether you actually need this level of details to answer.

I remember this recent question about a nuked gunsafe at point blank range. While the question asked for hard-science and didn't provide much, I still answered it. Indeed, following the principle of charity, there was no added detail which would have changed the answer : At point-blank no furniture can survive it (as proven by the top-voted answer). Therefore, adding details was just unnecessary to deduce the outcome.

This may make the tag mostly unfit for these specific closures since by default they allow a very wide range of answers. The needed details would be pretty much on-par with what is expected everywhere else, meaning no additional rule would need to be called 😊.

That the question can be answered under these tags, supposing details are given

Then, let's remember that oftentimes people add tags without reading what they mean. In the rare case a question under or would be clearly unanswerable under the tag's terms -like when magic is omnipresent-, I'd rather say the querent didn't actually want that tag. Back when they chose their tags it just looked sparkly and colorful ✨.

Otherwise, if it could be answered with these tags (and with the details), I'd close it for lacking details. But first on whether the tag should have been used, then on whether the question has enough details. Concretely, I'd introduce a doubt in comments with these kinds of question:

  • Do you know what the tag means?
  • Did you read what it requires?
  • Is your intention [one of the above intentions]?

If the answer tells us the tag wasn't needed, I'd retract my close-vote and remove the tag from the question. This would be part of the overall process of clarifying the question.


I think a strict applying of tags would just result in no one using tags. I see tags as more of a way of shaping the intention of the author, and that answers should bear that in mind without feeling strictly bound to it.

e.g. as Tortliena brought up with the gunsafe, the question was "Can a gunsafe survive a nuclear blast level attack". A perfectly answerable question, with the answer being "no", but the hard-science tag means answers should prove it, which they did a good job of.

But imagine the question had been "How can I make gunsafe in my world survive a nuclear blast level attack" with a hard-science tag. A strict interpretation of the rules could imply that the question can't be answered, is therefore bad, and should be closed, but who is that helping? I think it can still be handled as a valid application of a tag; it just means that the answer is "you can't, and here's why".

The author was hoping for a hard-science answer but the actual answer is that there is no hard-science answer. Answers could even tickle with science-based or science-fiction despite the tag, and still be legitimate. "You can't, here's why (the hard-science tag demands this explanation), but maybe if your world invents something with the following properties, a gunsafe made of that material would survive."

It gets them the science-based answer their question needs while informing them that there was no hard-science solution.

  • $\begingroup$ These tags have been around since the beginning. They're part of the fabric of the Stack. The rules have been clarified over the years to help people understand the differences between them, but they serve a long-standing purpose. Non-hard-science answers, for example, to a hard-science question are in threat of deletion because we expect the OP to be as mature about their expectations as we do respondents their answers. The goal of neither this Stack nor Stack Exchange is simply to answer anything that comes along to any degree of quality. That's what Reddit and Quora are for. $\endgroup$
    – JBH
    Commented Jan 5, 2023 at 16:26
  • $\begingroup$ @jbh I don't think we disagree on the fundamentals. But if someone asks a worldbuilding hard-science physics question, where I see a difference between Physics and Worldbuilding is that Physics could say "no, [science here]", where we would say "no [science here], but here's how you can fix it" and continue to help them build their world. Otherwise there is no point in the tags here and we can just redirect all of them to hard science stacks. $\endgroup$
    – JamieB
    Commented Jan 5, 2023 at 17:11
  • $\begingroup$ If what the OP wants is "No, but..." then the question cannot be tagged hard-science. That's the point of the tag. No buts. If we need a discussion about changing the fundamental nature of the Stack, that needs to be introduced as a separate question (and keep in mind that you can't vary from what the service owners want: one-specific-objective-question/one-best-answer). $\endgroup$
    – JBH
    Commented Jan 5, 2023 at 17:13
  • $\begingroup$ @jbh Then it seems like your argument is that the hard-science tag should automatically force a close and all such question should be redirected, as we provide no additional value over the science stacks for such questions. Which really is a valid view and a possible path forward. The gun-safe question could have simply gone straight to Physics, unless maybe we don't think they would answer, shall we say, "XKCD style questions". My argument is really that we must provide a unique twist or else we serve no unique purpose in these cases. $\endgroup$
    – JamieB
    Commented Jan 5, 2023 at 17:19
  • $\begingroup$ This is a tricky issue, but the truth is that it happens on any question, tag or not ^^. The "you can't, here's why" is a frame-challenge, something I told about there and I strongly recommend not "answering" that statement without an alternative. Those should be in comment. [...] $\endgroup$ Commented Jan 5, 2023 at 22:04
  • $\begingroup$ [...] Now, if the question shall be closed, by the strict interpretation of the rules... No, because there's no valid close reason which tells you can close questions with no possible answer. It's as easy as that. [...] $\endgroup$ Commented Jan 5, 2023 at 22:04
  • $\begingroup$ [...] One final note : The gunsafe question could really be answered with hard-science. And it actually has : Historical references, calculations, etc. I don't see how you found there hasn't been one ^^. $\endgroup$ Commented Jan 5, 2023 at 22:05
  • $\begingroup$ You're a bit close minded about this. The suggestion is that if a question asks for a hard-science answer, it should be expected to provide the hard-science information required to achieve that answer. A hard-science question about an orbit should be reasonably expected to provide statistics about the star, the planet, other bodies in the system, etc. If it doesn't, then the behavior we have always had, supported by Stack Exchange's preference to close questions quickly, is to close the question, because any answer is not better than the right answer. $\endgroup$
    – JBH
    Commented Jan 6, 2023 at 6:20
  • $\begingroup$ @Tortliena If "lack of a valid close reason" prevented closures, I don't think I'd have nearly as much to argue about in meta! Typically people throw catch-alls like "opinion-based" or "community specific reason". You'd think some people here were paid by the closure. I feel pretty justified in worrying about VTCs on questions that didn't violate any rule, since it happens daily. $\endgroup$
    – JamieB
    Commented Jan 7, 2023 at 20:16

You must log in to answer this question.

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