Can we restrict what kinds of identification questions are on topic?

Question

Recently there was a significant discussion about whether a question on the main site should be on topic. I very strongly believe that it is not a good question for the site and should not be on topic here, but the community thought otherwise. That got me thinking about what defines our site's scope with respect to identification questions. I'm making this post to share my reasoning for why I think identification questions like that one should not be on topic here and see if I can convince the community to agree to make this policy.

To be clear, when I say "identification question", I'm talking about a type of question that simply asks "what is [thing]", where the "[thing]" might be an object or event seen in everyday life, or an object or optical feature in a photograph of everyday life (i.e. not a photo taken in a science experiment), or rarely, a video of something happening. There's often no indication of any prior research by the asker, and in particular, the asker doesn't share why they think the thing they're asking about has anything to do with physics or astronomy (as opposed to, say, chemistry or biology, or maybe they just thought "hey, scientists know all kinds of random things"). Also, the question doesn't ask to explain anything. E.g. if the thing being asked about is an event or optical feature in a photo, the asker doesn't ask why the thing happens, they just want a name. Or if they're asking about an object, they don't (seem to) care about what it could be used for.

I think these questions are bad for the site because

• They're basically trivia questions which may or may not have anything to do with physics, so e.g. a random person is just as likely to know the answer as a random physicist or astronomer. That doesn't make the question bad, but I think it should mean that this site is not the place for it. I'd like to think that questions which are well-received here are those which are more useful to, and more answerable by the audience described in our help center (active researchers, academics and students of physics and astronomy) than to the average person.
• Also, these questions don't prompt us to give an answer that shares any science knowledge. And if a question can be fully answered without sharing any science knowledge, it raises the question of what it's doing here, since neither the question nor its likely answers will be related to physics or astronomy. (Often one can post an answer that shares some physics/astronomy knowledge by going beyond what the question is asking for, but that's true of a wide variety of questions which I think we would all agree have no business here.)
• A lot of these questions don't get much traction on the site, but the ones that do tend to be quite popular and often hit the HNQ, because visitors who aren't necessarily all that familiar with science can understand them. Tying into my previous point, that means a decent fraction of the questions that do the most work to represent our site are marginally or not at all related to physics or astronomy.

With all that in mind, what I'm proposing is that we make a policy that puts the burden on the asker of an identification question to show why their question belongs here, specifically. I'm proposing that identification questions, which ask "what is [thing]", must meet one of these criteria:

• Explain why the question has something to do with physics or astronomy specifically (this would cover identification questions for devices used in physics experiments as well as astronomical objects)
• Ask for some kind of physics-related explanation of the thing described, e.g. why it occurs (if it's a phenomenon) or possibly how it formed

Under this proposal, for example, all four of the following questions that Yly linked in the other meta post are fine because they ask for explanations:

• Strange ice found in my garden
• When separating an Oreo cookie, why does the cream stick to just one side only?
• Why do shadows from the sun join each other when near enough?
• Why are the wet patches on these floor tiles circular?

There are a few questions where the answer would have to do with physics or astronomy but the asker doesn't know that or doesn't have any reason to believe that. Those would be rendered off-topic under this policy. I think that's reasonable because the asker can easily be guided to modify their question to make it clear that it does have to do with physics or astronomy.

Answer 1

I sympathize with the intent of this post but I think it would be detrimental to implement the proposed criteria in practice.

1. People observing a phenomenon and then asking what it is don't know what it is. How are they supposed to know whether e.g. some optical phenomenon they see is an optical illusion more explained by biology and neuroscience (think e.g. about still images that look like they're moving) or a "real" thing that is explained by physics?

   Instead of requiring an explanation of why they think it's "physics", I think we should just stringently use the needs details or clarity reason to close questions where not enough information is given to determine uniquely the phenomenon happening. If enough details are given, it's either close-able because the answer is not based in physics or uniquely answerable (yes, this is a rare case where I do think closing a question based on its potential answers is justified). The stringently is important there - we should not be willing to play a guessing game about physics that might be involved.

2. The idea that they should ask for a physical explanation instead of just "what it is" is at first glance reasonable to me, but what's stopping us from simply assuming that someone who posts "What is this?" on a physics site is interested in a physics explanation of the "this" in question? Why do we need to close the question and force the asker to essentially always just replace "What is this?" by "What is the physical explanation for this?"? This strikes me as policing the formulation of the question more than policing its content. I think it's safe to assume that the overwhelming majority of people asking such questions on physics.SE will be interested in the physical explanation, and even if they're not - do we have to care?

   If the fear here is that if we don't do this then we get answers that just say "This is called X" without any explanation at all then I would like to believe that we'd just not upvote such answers (or even - gasp! - downvote them). If this belief isn't enough I'd prefer just making it a policy that such answers are considered non-answers (and hence deletable via low quality review) rather than requiring askers to jump through this specific hoop in how they're phrasing their question.