This question cs.stackexchange.com/questions/108496/… is literally p != np and check my proof at once. First is off-topic afaik, second is frowned upon. Do we have policy change?

@Evil I'm not aware of P vs. NP being off-topic explicitely. I've answered questions regarding that before and I haven't seen anyone making similar remarks. Regarding 'check my proof', it is usually only interesting if the proof is wrong. The combination of the two is in fact an improvement, because an elementary proof on P vs. NP is likely to be wrong.

@Evil Is there are specific discussion on meta that makes you think we have a policy against this question?

@Evil Or, otherwise, do you think we should do anything against this particular question? I think it is fine as of now.

@Discretelizard first: cs.meta.stackexchange.com/questions/786/… also I have seen comment based debates, general guidance that we are not peer review site and scientific advancements are beyond the scope of this site.

@Discretelizard it is borderline fine, and this bit of decision is a bit fuzzy, so I have asked to support any side

@Evil I do not really see any explicit advice against questions regarding P vs NP. More against long winded and unclear proof attempts. I have the feeling that that question does not consider an elementary proof. Also, that meta is pretty old (like many, but still) and the question is a bit vague, so I'd rather have new discussion than referring to that one.

@Evil Personally, I think the divide between "check my proof" and questions about why a certain argument does not work that I think are a good fit for Computer Science, is whether it is reasonable to extract a 'critical flaw' from the argument and that the argument itself is clear. Most "check my proof" question have a proof that is simply poorly presented, or where the critical error (or errors) are not clear.

Of course, this is all way too fuzzy and dependent on things the asker does not know to be useful as a general rule, but this is how I mostly judge individual cases.

@Evil another follow up seemingly cs.stackexchange.com/q/108538/68251