Result: Thanks to Dan Russell’s thoughtful answer and others’ constructive comments, the puzzle statement has been reupholstered, the puzzle has been reopened, and a complete solution is imminent.

Original post

The rainbow mystery below was readily understood by three posters and two other commenters as the puzzle it was meant to be. The three posted answers presented astutely-targeted partial solutions based on clues, well before “off topic” voting began and led to closure even though this puzzle was off to a good start with no symptoms of being off topic.

Question of consistency: Is there something about this puzzle that lands it in the same bin as a mathematics “problem” vs puzzle?

Bonus question: Should it have been presented differently?
Admittedly, less time was spent on wording than on the picture and I would understand downvotes and constructive comments rather than closevotes.

$\small\dfrac{ \raise-.8ex{\scriptsize+} \raise-1ex5 \, }0 ~$ Two many rainbows? [on hold]

Wish I’d had a camera at the time, but a cartoon will have to do.

This represents a direct view of two actual incomplete rainbow arcs that stop in midair where they cross, lit only by a setting sun. How could this be? I honestly wondered if it was a dream.

        What is the simplest explanation for this odd pair of rainbows?

                Why is the smaller one slightly brighter?

Notes. Each main arch was accompanied by a rainbow’s usual set of concentric fainter arcs, which exhibited the same phenomenon of stopping sharply where they crossed exactly above or below the main crossing point. Only air is between the point of view and the rainbows. Safe to guess that this effect never occurred more than a century or two ago. Some details of the real-life story have been altered in order to stymie internet searches.

[visual] [science] [real]

The following comment registered stupefyingly many upticks.

I'm voting to close this question as off-topic because this belongs on physics.stackexchange.com.
– closevoter with, ironically, no apparent affiliation to physics.stackexchange.com

At least the commenter was considerate enough to share their reasoning, but the confident tone may have misled others to take evaluation shortcuts as it might have given the impression that someone had already accurately assessed the puzzle. When this comment’s upticks correlated to closevotes, astonishment led to the impression that all closevoters combined for almost no apparent presence whatsoever at any science SE sites, finding a total of one comment and one all-but-ignored post at any such site, which was not Physics SE. Update: 4 of the upticks may have been automatically applied by the system without closevoters specifically endorsing the comment’s details.

Did any closevoter realize that essentially the same puzzle could have been stated without science, in terms of [clickable/hoverable hint /spoiler] for example? A pair of rainbows is just an especially intriguing manifestation of the solution, with natural clues that make it a better puzzle, and happens to be how the paradox presented itself in real life.

Did any closevoter genuinely imagine a complete solution that would verify that this is not a puzzle? The solution is quite simple and probably understandable by most solvers, but does hinge on a less-than-obvious aha-like detail from everyday experience.

Perhaps we could see specific reasoning posted here from those who voted to close, but please do not be tempted into defensive rationalization.

Mistaken closures based on hunches occur often enough that I openly suspect this to be merely another instance and hope that we can clarify a nebulous border between on- and off-topicality.

  • 6
    $\begingroup$ I remain in neutral territory, but just fyi, I think when you VTC with a comment, then when others select the same VTC reason, the system automatically upvotes the comment. $\endgroup$
    – Alconja
    Commented Oct 22, 2016 at 21:35
  • $\begingroup$ How would the system know which comment to upvote? At times there were earlier and later comments that were not upticked. $\endgroup$
    – humn
    Commented Oct 22, 2016 at 21:41
  • 1
    $\begingroup$ because the system made the comment too. A user selects VTC, and enters a custom reason, the system add that reason as a comment, on the user's behalf. (I think) $\endgroup$
    – Alconja
    Commented Oct 22, 2016 at 21:43
  • $\begingroup$ Interesting, but in this case the comment was posted minutes before the initial close vote. Would indeed make a difference to know if these upticks were automatic. $\endgroup$
    – humn
    Commented Oct 22, 2016 at 21:44
  • $\begingroup$ perhaps I'm just outright wrong then. :) Either way, your point (re: people taking stock in a confidently worded comment), is still valid. $\endgroup$
    – Alconja
    Commented Oct 22, 2016 at 21:48
  • 6
    $\begingroup$ Alconja is correct. When a user selects a custom close reason, they type "I'm voting to close because blah blah blah" into the question-closing popup, which then gets automatically added as a comment under their name. If a later close-voter selects the same custom close reason (e.g. from the review queue), then an upvote is automatically added from their account to the comment. That accounts for up to 4 of the upvotes on Ian's comment; the remaining upvotes must have been just from people who agreed with the comment without voting to close. $\endgroup$ Commented Oct 23, 2016 at 1:01
  • 2
    $\begingroup$ I didn't vote to close, because this is a remarkably amazing, and yet confusing, puzzle :) $\endgroup$
    – ABcDexter
    Commented Oct 24, 2016 at 5:40
  • 1
    $\begingroup$ I would've kept this open as well - I think the catalyst for the close votes was the phrasing that made it seem like any ordinary Physics.SE question. $\endgroup$
    – Deusovi Mod
    Commented Oct 25, 2016 at 18:10
  • $\begingroup$ I still think this question belongs better on physics.SE. It seems that you're asking us to explain some natural phenomenon, and in my opinion if that were the case I wouldn't really consider that a puzzle. $\endgroup$ Commented Oct 27, 2016 at 0:19

1 Answer 1


How I'd answer your questions:

Question of consistency: Is there something about this puzzle that lands it in the same bin as a mathematics “problem” vs puzzle?

Sort of. I admit when I first read it, I thought you were merely asking us to explain a natural phenomenon you'd observed, which I think is why some people thought that Physics SE was a better place for it.

Bonus question: Should it have been presented differently?

Yes. Because (other than obviously it being posted on Puzzling) it's not clear that this is a puzzle rather than a request for an explanation. I think the only additional thing it needed was a little explicit statement that though this may seem like a simple physics question, it is indeed a puzzle with some sort of puzzly twist. I think there's nothing wrong with this type of direct communication. See, for example, this recent question from @GentlePurpleRain where he anticipates then heads off some potential objections by simply stating "I'm aware of what not to do. This is tagged correctly."

All that said, given your very positive track record here and the fact that it's not obviously just a physics question, I wouldn't vote to close it. I think in general: When in doubt, don't VTC. And I've voted to reopen.

  • $\begingroup$ These suggestions will indeed improve and clarify the puzzle statement. Live and learn, with help. $\endgroup$
    – humn
    Commented Oct 24, 2016 at 18:39

You must log in to answer this question.

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