@TeresaLisbon Wow! I learnt something new today. That’s an interesting mistake to make, but an upvoted and accepted answer is not the best place for it to exist. I have downvoted the answer.
Why $a=x^{\frac{1}{m}}+\epsilon_0\mod(2\pi)$ implies $a^m=(x^{\frac{1}{m}}+\epsilon_0)^m\mod(2\pi)$? For instance, $0=2\pi \mod(2\pi)$, but $0^2\ne (2\pi)^2 \mod(2\pi)$. — Alex RavskyJul 18 '20 at 5:21
@Peter Close as what? I mean, I think it's not a bad thing to discuss. Maybe it can be made more on-topic in the following sense : devising an algorithm that takes very little input and running time and is capable of outputting a sufficiently random coin toss. That feels more on-topic and relevant. I would put the answer as off-topic to the question, but not the question itself because it's closer to what I asked, and IMO what I've asked for should be on-topic.
I think the question can be classed as unclear , because there is no initial source of randomness, and picking one out of real-life is a question that wouldn't probably be appropriate for MSE.
Usually, we talk about generating randomness FROM an already present source of randomness. If that source isn't specified or is a real-life object then IMO "what" real-life object is sufficiently random could be opinion-based.
@Peter I'll take down the C227 when I can. By the way I'm commenting on the question you attached, but if anything I am probably going to close as opinion based right now.
@Peter I've voted to close that question as "opinion-based". I expect people to cry foul but this is definitely the case from my comment.
@user21820 I'm sorry : I just wanted to save people's time by stating that, rather than have each one read it. I should have avoided that. It had 6 upvotes when I saw it first, so thanks for bringing it down to -1. I noticed this while looking for an easier proof of sin(n^2) being dense in [-1,1] and spotted the mistake then. Sorry.
The answer is gone now.
@Peter It got closed as a duplicate, but the original question is then quite opinion-based!
@TeresaLisbon No problem! I understood your intention (to save people's time). But to avoid giving others the wrong impression, you could instead pinpoint the particular step, or in this case highlight the existing comment about "0^2 ≢ (2π)^2 (mod 2π)".
@user21820 Sorry, I agree I wasn't precise enough when saying "the exponentiation step" but what I meant was that $a = b mod 2\pi$ doesn't make $a^k = b^k mod 2pi$ for integer $k$, and that step was wrongly used. Thankfully, it's gone now.
@XanderHenderson I'd lean on the side of rapid closure; but I get that the answers are worrisome. Maybe you could write a comment to the OP, or a CW answer, before closure. But I'd have no problems with voting to close it.
Another declined flag about a comment about this question :
"The obvious approach is to use a Computer Algebra System such as Wolfram Alpha." flagged as "not relevant to the post". To my opinion, a ridcolous "declined".
@Peter The question is overly broad, and it is not clear that the comment really is irrelevant (I mean, for anything really painful, I tend to fire up Maple just to check my work, so "use a CAS" isn't really an unreasonable answer).
While I am not the moderator who cleared that flag, I don't disagree with the action that was taken.
@XanderHenderson Closed it, answers don't influence me enough in this case, and likely a duplicate as well (complete the square-type).
@XanderHenderson It reopened actually, now has three close votes. One for the "close, don't delete" functionality, but certainly not open, in my opinion.
@XanderHenderson I'm not sure, either. It's clear the OP put in effort, but, no one should enable this op thinking, "Oh, great, all I need to do is link copies of my work." I commented below the answerer who praised the OP for the image of their work. It's tough. Perhaps leaving a comment with a link to the supposedly "short briefer on math.jax"?