CURED - 2020-11-19

KReiser

00:10

no effort, close please

amWhy

02:21

This needs to be closed, asap

Martin Sleziak

09:08

duplicates I voted to close Show that $f$ is convex iff $f\left (\frac{1}{2}x + \frac{1}{2}y \right ) \leq \frac{1}{2} f(x) + \frac{1}{2} f(y)$ as a duplicate. But since it contains two questions, maybe somebody would be willing to add a close vote with another duplicate target.

Namely this one: Example of a function such that $\varphi\left(\frac{x+y}{2}\right)\leq \frac{\varphi(x)+\varphi(y)}{2}$ but $\varphi$ is not convex

For the claim about continuous functions, see this question: Midpoint-Convex and Continuous Implies Convex (and other questions linked there). For the counterexample, see this question: Example of a function such that $\varphi\left(\frac{x+y}{2}\right)\leq \frac{\varphi(x)+\varphi(y)}{2}$ but $\varphi$ is not convex (and other questions linked there). — Martin Sleziak 14 hours ago

@MartinSleziak that should be an answer in my opinion. — supinf 14 hours ago

@supinf I am not really sure about that. I was considering voting to close as a duplicate, however, there is a problem that there are actually two duplicate target. (Moreover, another difference is that this is about convex subset of $\mathbb R^n$ and the linked questions are about real functions - but I do not think that the change between $\mathbb R$ and $\mathbb R^n$ changes the proof too much.) — Martin Sleziak 14 hours ago

Does this answer your question? Midpoint-Convex and Continuous Implies Convex — Martin Sleziak 1 min ago

And, of course, if you think that it should not be closed as a duplicate (or that there is a better duplicate target for a questions dealing with $\mathbb R^n$ rather than $\mathbb R$), let me know.

José Carlos Santos

@MartinSleziak I've closed it as a duplicate with two targets.

Martin Sleziak

Yes, I have just noticed: math.stackexchange.com/posts/3913068/revisions

Thanks for doing that!

SmokeDetector

09:32

[ SmokeDetector | MS ] Bad keyword in username (89): limits to infinity at a point by Keto on math.SE

Arctic Char

10:04

D1, D2, D3, D4

D5, D6, D7, D8

D9, D10, D11, D12

C1, C2, C3

KReiser

11:00

no context, close please

Peter

11:43

-7

Q: They weren't being intentionally misleading...right?

Offset linear PDE I don't see either of them actually solving the problem and verifying their claims that it's solvable. Offset PDE are not something I see often, I only occasionally see something tangentially related to them when an author happens to discuss a particular linear solution or somet...

partial-differential-equations

Peter

12:03

-4

Q: Result of divison by Zero

When we divide by a number then the result represent that the dividend can be divided into quotient in divisor ways. For example like 15/3=5 means 15 can be divided into 5 people equally in 3 ways. so if we divide something by zero like 15/0 then does it mean 15 can be divided into 0 people in in...

combinatorics divisibility

Another division-by-0 question

Xander Henderson

del-pls 20k+ algebraic-topology Link-only answer; the link is dead. Answer to: Hopf map fiber bundle help‭ - Neal‭ 2012-04-21 13:46:35Z

I don't want to delete the above answer, as my deletion would be permanent, and the original answerer might show up to improve the answer.

However, it probably should be deleted.

Paramanand Singh

12:54

@XanderHenderson: I don't see the delete button. Do i need to downvote first?

Martin Sleziak

From the FAQ: How does deleting work? What can cause a post to be deleted, and what does that actually mean? What are the criteria for deletion?

> Users with reputation ≥ 20k (more precisely, the trusted user privilege; 4k on beta sites), can vote to delete closed questions within 48 hours of being closed, so long as they score −3 or lower. They may also vote to delete answers of score −1 or lower (0 or lower if through the Low Quality Posts review queue), including accepted answers (unless they are the owner of that answer). It takes three votes to delete an answer.

@ParamanandSingh AFAICT, only answer with negative score can be deleted by votes. (I wasn't sure whether accepted answer can be deleted by voting - but the FAQ says it is possible.)

Paramanand Singh

14:17

Thanks @MartinSleziak for the info.

Xander Henderson

14:30

@ParamanandSingh Oh, I had forgotten that answers needed to have negative score to be deleted.

I've dealt with it.

amWhy

14:56

@XanderHenderson @Xander, what happens in the low-quality RQ, when a reviewer has the option to delete a poor answer, even when at score = 0? Are they seen as "recommend deletion" only?

Xander Henderson

@amWhy I have no idea. I'll bet @MartinSleziak knows, though.

amWhy

@XanderHenderson You don't know everything about math.se and SE?? But you're a moderator! :P

Paramanand Singh

15:29

@amWhy: that kind of knowledge is possessed only by a few like Martin Sleziak and Glorfindel.

Martin Sleziak

15:59

While this is flattering, I definitely know much less about Stack Exchange software than Glorfindel does. And review queues belong to aspects of SE that I am less familiar with.

I think that in this case id doesn't matter what is the score. It is taken as deleted (for 20k+ users, 10k+ users can only choose recommend deletion).

Looking at the FAQ post on review queues, it seems that the post can be deleted even if it has score 0: "if a sufficient number of reviewers all recommend deletion (6 on most sites, 4 on Stack Overflow), it can cause the post to be either deleted (if the score is less than or equal to 0) or forwarded to moderators for further review."

Although I will certainly admit that the wording there is not entirely clear.

Martin Sleziak

16:18

I thought that it would be possible to get the list of scores of deleted answers which went through the LQP review queue from SEDE, but it seems that the ReviewTasks table does not contain the information on deleted posts: Can the reviewtasks table include the reviews for deleted posts as well?

So I don't know an example at hand. And I do not know some quick way to find such example. :-(

Xander Henderson

16:30

@Glorfindel Knows everything about the inner workings of the software. I hope that he doesn't mind that i just pinged him here...

Back to work...

Glorfindel

@XanderHenderson no problem.

Here is a 0-scoring answer deleted by three 20k users (or equivalent) from review:

boardgames.stackexchange.com/review/low-quality-posts/29041

(the first review is by a non-'trusted' user, but that doesn't matter)

Finding an example on Math.SE won't be easy, somebody with sufficient reputation could check the review history. I just happened to know one site where we regularly delete 0 scoring answers.

Martin Sleziak

17:39

3×delete, 1xrecommend deletion, score 0: math.stackexchange.com/review/low-quality-posts/1491439

2×delete, 4xrecommend deletion, score 0: math.stackexchange.com/review/low-quality-posts/1491592

3×delete, 1xrecommend deletion, score 0: math.stackexchange.com/review/low-quality-posts/1491700

I hoped that there might be a cleverer way to find them then looking at past reviews one by one. OTOH they are not as rare as I thought, so it's not that many posts to go through.

amWhy

19:09

@JoséCarlosSantos I was wondering if you'd like to work with me on a proposal for an FAQ post on our meta wrt the Suggested Edit review queue, geared to users suggesting edits. I brought this up in "Whatever quid" with quid. There are others who could help out, perhaps through a temporary chat room. I don't want to act like I'm the only one with a lot of input on the matter. See the discussion with quid:

in whatever, quid, yesterday, by amWhy

@quid I've been thinking that maybe we should consider a meta post or a meta FAQ addressing suggested edits: E.g., if the only edit you plan to make is capitalizing the first letter of the first word in a title, that's not an improvement, unless other real formatting issues or grammatical issues, and typos e.g., exist AND you also edit those. E.g., Adding a random space (e.g., Changing "If $x\in \mathbb R$ and $x\gt -2$, then $$\text{foo-equation}$$ to
"if $$x \in mathbb R$$ and $$x\gt -2$$ then $$\text{foo-equation}$$. (That sort of suggested edit happens **all the time**.) Of course w…

^^^ Ditto @Xander, @ParamanandSingh, and anyone interested!

José Carlos Santos

19:29

@amWhy I have been accepting those suggested edits, but I agree that they don't make much sense. Of course, if many users start to reject those suggestions, it will be better for all. So, yes, I'm in.

amWhy

@JoséCarlosSantos Great! I'll keep you posted.

epimorphic

Closure of math.stackexchange.com/questions/530639/… as duplicate seems wrong to me. Both OPs make it clear that what they're asking is distinct from the other person's question (counting measure vs scaling to uniform probability measure), and all the answers follow that separation. You also can't go from one result to the other by scaling post taking the p→0 limit.

amWhy

Please consider voting to close

José Carlos Santos

20:19

delete Bad PSQ

amWhy

Please consider deletion

Please delete. Look at the non-question, not the upvotes.

Xander Henderson

21:15

@amWhy Is there a good dupe target for that question? I honestly think that it is a good question to have on the site (though the particular post is terrible). If you can't find a good dupe target, do you have any interest in rewriting it?

If not, I might get to it in a few days.

amWhy

@XanderHenderson I'll take another look. I can't do anything yet tonight. Do you mean by rewriting it: posting a good, new question to that post? I do not condone editing posts to put words in the mouth of the asker.

Xander Henderson

@amWhy Yes, I mean posting a new question, in accordance with the rewrite policy posted a couple of months ago.

I did not mean to imply that you should edit the question.

amWhy

@XanderHenderson I thought so! Just wanted to make sure.

Xander Henderson

Honestly, I imagine that there is a duplicate version of the question out there somewhere.

It seems like a natural question to ask (and gets at a heuristic argument for the prime number theorem).

So I would like to see the good answer preserved, but that question is simply not up to snuff.

amWhy

@XanderHenderson Indeed. If a new/better version were posted, the good answer can then be migrated?

Xander Henderson

21:22

@amWhy My hope would be to merge the old question into the new. Yes.

amWhy

@XanderHenderson There's a clear pattern that good questions highly upvoted get lots of answerers, or a few, all highly upvoted, and vice versa, a great answer with lots of upvotes seem to trigger users to upvote the question as well. Clearly, highly upvoted questions and/or answers tend to lead to higher viewership. But it is still rather puzzling: "the transitivity of upvotes".

The question can be phrased, also, "Let a and b be random independent positive integers. What is the probability that the $\gcd(a, b) = 1$.

Xander Henderson

"The transitivity of upvotes." I like that.

amWhy

@XanderHenderson ;D

Transcript for

$Mathematics$ CURED