CURED - 2019-12-02

user12692

00:03

@BillDubuque But we are talking about general policy, neither are we talking about tons of questions depending on parameters, right? We are talking about this specific case.

user12692

I very much agree with you that we should be "guiding students to the most insightful answers." And hence my previous question about how we would need to guide someone who need help with $x^2+35=7^n$ to identify the one you linked. Neither Google nor Approach0 returns satisfactory results so far.

Bill Dubuque

@Jack Many of these questions have the parameters changed every semester. So specific parameter searches in those cases aren't going to be helpful. And, again, almost all askers are lazy and don't search. We end up having to search (for dupes). And I have wasted much valuable time having to wade through tons of duplicate answers that should have been deleted long ago.

@Jack You are talking about a specific case. I am talking about the general issue (which is far more important).

user12692

00:49

@BillDubuque Well, I fail to see the general issue applies in this specific case. You do not seem to have an idea what kind of search for $x^2+35=7^n$ would lead to your abstract dupe either, do you?

Bill Dubuque

02:14

@Jack As I said, it's a moot point since askers rarely search.

In any case, if you think those particular parameters will be searched for (unlikely) then you could simply add them to the body of abstract dupe question (e.g. "some other common instances are ....").

Yanior Weg

08:44

Duplicate: math.stackexchange.com/q/3457119/407165

Martin Sleziak

@BillDubuque I think that the point Jack is trying to make is that even when you search for a different version of this problem (different names of variables, different numbers) in Approach0, then you would get this post among the search result. (If it's not delete - and when it's eventually indexed.)

This is precisely the advantage of search engines such as Approach0 or SearchOnMath. They are good for searching formulas.

If you search using Google or the on-site search, then searching for x^2+y^2=z^2 does not return a^2+b^2=c^2. Change of one number or change of variable name removes this from result.

In such cases, it is better to use for searching some search engine which is able to recognize variables.

A situation is different when you're searching for something which has a name or you have good keywords that very likely appear in the post. Or if you're searching something which is very likely in the frequent tab of some tag. Then on-site search (or Google) might be better option.

Also saying that "askers rarely search" (even if this is true) isn't a reason to make things more difficult for people who try to search on site and to find duplicates. As mentioned several times on meta duplicates can be useful - typically if the formulation in one of them is different enough so that it might help somebody finding the post, if the user chooses different version of the formula or different search terms.

I will also add that, as far as I can tell, if a post is edited and it already has been indexed, Approach0 will work with the old version: chat.stackexchange.com/transcript/message/52758714#52758714

We'll see whether this will change eventually - if the Approach0 website grows.

Martin Sleziak

09:14

To include just one example, there are quite a lot post about showing $\prod_{t = 1}^{\infty} (1-x_{t}) = 0$ under the assumption that $\sum x_t=\infty.$

When I search in Approach0, I get some post containing the expression I searched for. (Some of the with different name of variables.)

Even if I don't have the best post among those results or if some of them are duplicates, after I find them I can chek the posts which are linked from there.

Martin Sleziak

09:31

I'll just add that it is perhaps not that important what happens with this specific question. (We'll see whether the mods will intervene if there are several rounds of deletion and undeletion, but so far there was only one.)

But as a general principle, it is good to do things that help with searching. (Such as, for example, correct tagging of questions. Or leaving a duplicate on the site if the formulation of the problem is sufficiently different and it might help people who are searching for the post using a different search query. (Of course, assuming that the post is not of too low quality - improving search does not negate other reasons why post might be deleted.)

user

@XanderHenderson It seems that the thread was removed even if I didn't flag it. Anyway, the case is now completely closed from my side. Sorry for the misunderstanding if so. Thanks

Bill Dubuque

14:43

@MartinSleziak Of course I know well the advantages of formula aware search engines (I have even worked on such long ago - much more powerful than approach0). I've already addressed above the points that you raised.

Martin Sleziak

@BillDubuque Well, then I don't really understand what you were trying to say actually.

You claimed that if parameters are changed, then it's impossible to find such post.

Sorry for the typo - and the double ping.

Do you at least agree that it actions which make it easier to search for posts about some problem are useful for the site?

Bill Dubuque

Btw, a matter of ethics worth emphasis. One of the users (user=gimusi) arguing for undeletion has a highly upvoted answer (+6) in this post and has not disclosed that fact. Not the first time that user has been accused of dubious ethical practices in posts involving his answers. If GENTLE users support such behavior then I will no longer support them (even though my views are closer to GENTLE than to CRUDE).

Martin Sleziak

I put usefulness for site above issues concerning reputation. (And, if I remember correctly, you have also said that you don't really care about reputation.)

Bill Dubuque

@MartinSleziak I don't understand the point of your prior comment.

Martin Sleziak

14:58

Although the objection that if somebody is for undeletion and gain something from it, then they should disclose it is fair.

@BillDubuque The one that you replied to?

Ok, so probably that one.

Bill Dubuque

@MartinSleziak You are making more unfounded inferences. I raised what I believe is a matter of ethics of concern to many members of this room. You seem to think that implies something about my personal views about reputation, etc. Non sequitur inference (and a topic which will only serve to divert the discussion onto nonconstructive tangents).

Martin Sleziak

Ok, is the fact that gimusi did not disclose that they have an answer there the only reason why you think that this particular post should be deleted?

Bill Dubuque

@MartinSleziak Again unfounded & nonconstructive. Why in the world do you believe that someone thinks that?

Martin Sleziak

I am simply asking whether this is the case.

Hopefully, since gimusi read your objection, they will mention in the future whether they have an answer on the question they are asking to undelete/reopen.

So from your answer it seems that there are also other reasons why you think this post should be deleted, right?

So it seems that you think that having this post on the site is not useful, while I think that it is useful. It seems unlikely that we're going to agree - so why don't we leave it at that.

user

15:15

@BillDubuque I always submit to GENTLE questions to which I've aswered for the simple reason that those are questions I know and I can say if they deserve closure/deletion or not. I also have submitted questions with downvotes to my answer, it is not the main issue for me. Moreover I think that when upvotes are high the deletion do not affect the reputation, but I'm not really sure on that.

Xander Henderson

@AlexanderGruber Good morning! (or evening, or night, depending on where you are)

Alexander Gruber

@Jack Thanks, good suggestion. Dropped a 5er on it

Morning @Xander! I'm in Chicago now

Martin Sleziak

@user Actually, only answers that were on the site for 60 days before the deletion and have high enough score are immune to reputation loss from deletion. See here: Do you lose the reputation (points) you acquired from a question if/when it is deleted?

Alexander Gruber

Which is turning out to be kind of the opposite of Florida where I was until last month

Bill Dubuque

@MartinSleziak Again NO - please stop making wild guesses unsupported by facts. I explained above why I believe such dupes should be deleted and that explanation has nothing at all to do with the fact that "gimusi did not dicslose they have an answer". How you could believe something so bizarre is jaw dropping.

Martin Sleziak

15:19

Fair enough. Anyway, I have simply asked a question - without guessing whether the answer is yes or no.

I will repeat what I said above: it seems that you think that having this post on the site is not useful, while I think that it is useful. It seems unlikely that we're going to agree - so why don't we leave it at that.

Xander Henderson

@AlexanderGruber Chicago is a great city, but late fall or winter is not the right time to be there if you want to go outside. :)

And with that comment, I'm going to go back to not making off-topic comments.

user

@BillDubuque I'he just "disclosed" that almost all the questions I submit to GENTLE contain an answer given by me. Moreover anyone can see directly by the linked OP this fact. I think that there is not an ethic issue on that, since all is so transparent. Thanks anyway for the suggestion.

Martin Sleziak

Well, I'm single.

Bill Dubuque

@MartinSleziak Did you actually intend to mislead others to make them believe that I said that?

Martin Sleziak

No.

Bill Dubuque

15:26

@MartinSleziak Then why did you write that?

Martin Sleziak

I asked whether or not this is the case.

You were the one who raised the issue about reputation.

user12692

@AlexanderGruber That's great. Thanks!

Martin Sleziak

Whoever flagged it: I don't think Bill's message was rude.

And now the situation is even worse - Bill was trying to respond to my message, his response is now deleted.

If you believe that there was some malicious intent behind the message you objected to, I don't really have any way to persuade you that's not the case.

In any case, you have answered clearly that the answer to my question is "No."

Oh, I see that Bill has been actually blocked from chat - I did not notice that until now. That's rather unfortunate.

Bill Dubuque

16:07

For the record, my above mod-deleted comment said "But your "question" was loaded - you asked if gimusi's non-disclosure was "the only reason I think this post should be deleted". Most readers will wrongly infer from that that I said it was one reason - when I never said anything of the sort. Have you stopped beating your wife yet?"

The final "beating" sentence is the prototypical English example of a loaded question (google it). It seems some nonnative English speakers didn't realize that and took it literally - not realizing the context and the intended point. Sigh.

user

16:28

@BillDubuque Maybe for that reason our CoC suggest to "Avoid sarcasm and be careful with jokes — tone is hard to decipher online". Indeed for nonnative English speakers it is really hard to follow these kind of discussions. Maybe when we talk about deletion/closure issues we need to take into account this fact.

Bill Dubuque

@user It was neither sarcasm nor a joke. If you participate in moderating an English language forum then I think a requirement should be to have a good command of the language. It's hard to imagine that anyone would interpret that literally. You can always google strange-looking phrases to learn their meaning.

user

16:57

@BillDubuque Thanks for the suggestion, I try to do my best with the comprehension but it is really not always a simple issue for nonnative English speakers. For that reason I also suggest to all native English speakers, when possible, to avoid complicated formulations of their ideas or at least indicate the usage of strange-looking phrases. Thanks

user21820

17:30

D1, D2, D3.

D4, D5, D6.

D7, D8, D9.

D10, D11, D12.

D13, D14, D15.

C1, C2, C3.

C4, C5, C6.

@BillDubuque: You may want to look at this question and one questionable answer.

Bill Dubuque

18:01

@user21820 D16 I already handled that, and the thread - a dupe of huge FAQ (casting nines) - needs only one more delete vote.

Bill Dubuque

18:26

@user21820 I wish others would help out with such "handwaving" arguments that often border on pseudo-mathematics. It looks very bad for the site to allow such content to stand. That user is notorious for such (besides the answers, there is also much posted in comments and this is hard for us non-diamond users to handle)

user21820

18:38

@BillDubuque You mean like this?

Bill Dubuque

@user21820 I hadn't seen that, but it is - alas - far from the worst I've encountered.

Jyrki Lahtonen

18:54

@user Users simply should never advertize threads they have answered. That counts as rep farming. It is ok, if there is a motive to get something else closed as a duplicate. But in that case the link should be accompanied by a disclaimer.

2

user

20:32

@JyrkiLahtonen I'm not advertizing indeed, I simply submit publicly to the community the questions for which I believe that closure and/or deletion are not motivated. I can't see any misconduct in the submition, anyone can see who answered the questions and decide franky whether questions deserve to be reopened/undeleted or not.

@JyrkiLahtonen Rep farming is not a point, indeed I think that when a question is deleted whitout motivation, the undelation, if the case, is a fully legitimate Rep earning. We don't steal rep points, we receive that from many other users which appreciate our contribution and they thanks me, or any othe users, in that way.

Jyrki Lahtonen

21:01

@user You are one to talk, when 90% (or other such high fraction) of your rep is "earned" by answering dupes. Often you are self-plagiarizing, and then you have the gall to advertize those posts, begging for others to undelete them.

5

user

@JyrkiLahtonen Your way to see things puts all in a very negative perspective to me. I'm sorry for that but I'm aware that we have also a very different perspective for the site. I don't understand but respect your point of view.

Bill Dubuque

C1 was undeleted and reopened mostly by "GENTLE" users, and by a user with a highly voted answer there (gimusi) I will no longer work with GENTLE due to such destruction of site organization.

user

21:18

@BillDubuque Sorry but I can't see what is the point with the fact that the I have an answer there. Does the question deserve closure or deletion? We should discuss of that here and on Gentle and not who have answered. Personalization are always not a good thing.

user12692

@BillDubuque I do not think you have ever "worked with GENTLE".

user12692

@BillDubuque And do you make disagreement with your decision as "destruction of site organization"?

Bill Dubuque

@Jack Then you're wrong about that too, just as your math is wrong here.

user12692

@BillDubuque If you do not agree, you could simply say so. As I am not convinced by your argument for deletion, I did not, would not, say that you are "destructing the site". On the contrary, I firmly believe that your deletion of this post "is making this site unnecessarily difficult for searching".

Bill Dubuque

@Jack I already debunked that weak searching argument above. One could make the same claim against almost every deletion.

user12692

21:26

19 hours ago, by Bill Dubuque

In any case, if you think those particular parameters will be searched for (unlikely) then you could simply add them to the body of abstract dupe question (e.g. "some other common instances are ....").

user12692

Perhaps we should do that (add them to the body of abstract dupe question (e.g. "some other common instances are ....") first to see if one can find the post.

Bill Dubuque

@Jack It's your argument to attempt to support - not mine. I already explained why I think this is a waste of time.

user12692

@BillDubuque I do not follow.

user

@Jack What I really can't understand is the tremendous idiosyncrasy for similar questions or answers. We should look at the quality 10 good questions/answers are always better (or not worst) than 5 good questions/answers.

Bill Dubuque

@user Please do elaborate on this "tremendous idiosyncrasy". I have no clue what you refer to.

user

21:35

@BillDubuque It seems to me that often we focalize too much our attention of the number of similar questions and answers. What is the issue with the "C1" you have posted?. It seems a good question to me, there are good answers. If you have something to add just give your own answer, put the link you like and any other information you think can be useful. I can't see any problem with that.

user12692

@BillDubuque You are free to vote, so am I. I appreciate your effort of organization, but I find it very difficult to understand your decision of deletion. If the "organization" means making this site better, then I believe we all agree that such "organization" should make a post easy to search.

user12692

Since @JyrkiLahtonen has mentioned several times one should try to search using Approach0 to identify duplicates before answering, which is indeed a good idea, I am wondering how he would search in this case to identify the claimed "dupe".

Bill Dubuque

@user The answers to that question are extremely poor - in terms of both rigor and pedagogy. See my comments there.

e.g. "it is easy to see that" (that particular inference is one of the most crucial in all of number theory. i.e. using Euclid's Lemma or FTA and one should never justify it by such handwaving)

And your answer is even worse leaving no clue what is intended. Is is not math. You'd get zero credit in my number theory course if you wrote that on a test.

user

21:58

@BillDubuque I think you should add all these observation as an answer in order to share these information with the asker and future readers. You could also give the link to more general questions in such way that in teh future other askers can easily find that. I don't see how closing and deletion can help to share these informations.

Bill Dubuque

@user I already did that in the linked dupe, which contains a complete and rigorous proof of parametrized generalization in the Lemma.

@Jack As I said numerous times above, if you sincerely believe that those specific parameters will aid searching then you can explicitly state that instance in the dupe. That is not a good argument for massive duplication of (poor) answers

RRL

22:16

Delete: DA, DB, DC, DD, DE, DF, DG, DH

user

@RRL I don't agree with many of these proposal.

RRL

22:34

@user: Which ones do you agree with?

darij grinberg

22:48

from a quick look, most of these have good answers that can help learners

DF is an open-and-shut merge candidate

user

@RRL Maybe DE even if I would give more time to the asker in order to improve its question.

Xander Henderson

23:27

@user If it is eligible for deletion, then the asker has already had two days (minimum) to fix their question. What makes you think that more time is going to lead to any improvement?

Bill Dubuque

23:37

Jyrki's critique is spot-on: all the reopen/undelete requests that user(gimusi) makes are in questions he answered, e.g. from the last few days:
[C84](https://math.stackexchange.com/questions/3448873/calculate-lim-n-to-infty-frac1-times-5-times-9-times-dots-times-4n-33)
[D85](https://math.stackexchange.com/questions/3445934/how-to-integrate-function-such-that-an-answer-isnt-infty)
[D86](https://math.stackexchange.com/questions/3441188/sum-of-series-sum-n-1-infty-left-fracnn-cdot-xn-right)
[D87](https://math.stackexchange.com/questions/3440989/how-to-argure-that-a-given-bivariate-function-is…

user

@BillDubuque I've already openly disclosed that here. There is nothing to hide with this. I submit questions for which I think closure and deletion are not motivated. I don't understand your point.

Bill Dubuque

@user Do you believe that every question you answer should not be deleted?

user

23:56

@BillDubuque I think that only a small percentange of questions should be deleted, when they are very poor and not improved by the asker at all. Otherwise I don't see any problem to keep good questions with good answers.

Transcript for

$Mathematics$ CURED