I have been put off by math.stackexchange's policy on people posting questions from ongoing math contests. A very recent violation occurred here: Triangle Geometry Question and I believe moderators will not divulge IP addresses of those who abused the system.

As a contestant on the Online math open I feel angry at inaction. I raised the question here: math.SE policy on question from ongoing contests and I am not happy with this.

I request the mod to delete my account on this site (you may keep the questions as you please) as a mark of solidarity for all my fellow high school students who compete in the olympiads and work hard for them. I also subscribe to self imposed standards of intellectual honesty and I am afraid by using math.stackexchange.com any further, I am collaborating with a website which does tolerate cheaters. Thank you. Goodbye math.stackexchange.com.

share
14  
Just FYI, the moderators are not allowed to make private information like IP addresses public even if they wanted, as they are bound by a moderator agreement. The SE privacy policy forbids them this. –  Mad Scientist Sep 27 '12 at 20:17
6  
Too late for None, but once your account is deleted, you have forfeited your ability to advocate change. –  robjohn Sep 27 '12 at 21:15
5  
It just occurred to me to wonder whether anyone has ever made a request for an ego or superego to be deleted. –  Gerry Myerson Sep 28 '12 at 22:38
7  
@GerryMyerson on other parts of the internet I frequently find myself wishing that some egos could be deleted. Or at least deflated –  user16299 Sep 30 '12 at 9:41
    
possible duplicate of Please delete this account. –  Andres Caicedo Sep 4 '13 at 2:47
    
@AndresCaicedo: other way round; that question would be a duplicate of this one. –  robjohn Sep 5 '13 at 10:56
add comment

4 Answers 4

In the comments to Qiaochu's comment, there is a dispute about whether moderators should delete/close questions from ongoing contests rapidly before they can attract answers. For example, Bill Dubuque writes "moderators should not execute unilatateral deletions on such contentious matters..." while Jonas Meyer writes "we would want a moderator to be able to [delete/lock these questions] in a timely fashion...".

I would like to poll the community, in the standard manner.

VOTE THIS ANSWER UP TO SUPPORT the moderators rapidly deleting or closing questions from ongoing contests, without waiting to seek consensus from the broader community.

I strongly support such closures, and will probably post a longer rant about this soon. (This answer, as is the norm for answers-used-as-polls, is CW.)

share
1  
Existing copyright policy covers leakage of questions before a contest officially opens (if the problem text is unique, and if not, a solution is already published). For contests in progress, an instant-close policy actually helps those who cheat, by maintaining their information advantage over the non-cheaters (who could at least check the answers to level the playing field, if alerted by the contest organizers and if MSE kept the question open). Informing the contest organizers, and attaching an "Ongoing Contest" tag, are actions that do not restrict what can be discussed here. –  zyx Sep 30 '12 at 21:55
    
Rant posted sbseminar.wordpress.com/2012/09/30/… . (Off site, to avoid being overly discussionny.) –  David Speyer Sep 30 '12 at 22:02
    
I think closing is better than deleting when no answers have appeared. When answers have appeared, there is no good solution, but I think deleting is better than notification, as it hides the answer from cheaters who would google for it but not post a question. Most cheaters are extremely lazy. I can't imagine why keeping the question open could be better than closing it; more answers will only do more harm. –  David Speyer Sep 30 '12 at 22:05
2  
I'm confused about the convention on vote questions like this, I would have expected the convention was upvotes only. –  Noah Snyder Sep 30 '12 at 22:15
4  
I'm not sure how anyone can make a case that moderators should not have such power. We voted them into power. If they don't have these types of powers and must wait for consensus before doing anything, why even have moderators? Let's just do everything by consensus. If you don't like how a moderator's unilateral decisions go, then don't vote for them next time around. It is like any representative democracy. –  Matt Sep 30 '12 at 22:17
    
@Noah My understanding is that "upvotes only" is the standard convention; that's why I created two complementary answers. As long as no one is double voting, I suppose it doesn't matter too much. –  David Speyer Sep 30 '12 at 22:18
5  
To point out explicitly what @NoahSnyder is saying implicitly, the two poll questions have received a number of up and down votes, so more people have voted than have appeared to do so. Hopefully, no one is abusing the system by voting twice. –  David Speyer Sep 30 '12 at 22:19
    
In light of @Matt's comment, I think there is an ambiguity in both poll questions as to whether it means the moderators should wait for consensus on general policy (and implement the consensus mostly independently thereafter) or wait in every specific instance where a contest question is posted. If the questions are edited to clarify this, anyone who misunderstood can re-vote. I interpreted Bill Dubuque's quote introducing the query as about general policy, but Matt's reading of the poll question is about individual instances. –  zyx Oct 1 '12 at 16:20
    
@zyx I don't think there is ambiguity. The proposal is about contest questions. I was using hyperbole to make my point. –  Matt Oct 2 '12 at 17:07
    
@Matt, under both interpretations the proposal is about contest questions, and the hyperbole was clear. The poll can be read as about what moderators should do from now until a policy exists, or about what the ultimate policy should be. –  zyx Oct 2 '12 at 22:04
add comment

Those in favour of giving away IPs of suspected cheaters should imagine that such a policy would allow me to obtain the IP of anybody asking a question here: Simply copy-paste a recent question asked to a throw-away website that looks like an online contest and - as "organizer" of the "contest" - claim that the asker is cheating in this "contest" and demand his IP address. Privacy is an important good that should not simply be thrown overboard.

share
    
Giving out IP addresses would be a really bad idea for many reasons, but it doesn't matter if anyone here wants to do that anyway. The SE privacy policy prohibits it and the moderators are bound to their agreement with SE, they're not allowed to give them out. –  Mad Scientist Sep 29 '12 at 8:42
7  
Yes, I just wanted to clarify that there is also a reson behind that policy. –  Hagen von Eitzen Sep 29 '12 at 9:28
4  
You can also use this technique to get any question closed/locked/deleted if the mods are to enforce a policy of closing/locking/deleting "contest questions". –  mt_ Sep 29 '12 at 10:07
    
@mt_ That sounds far fetched. Setting up a site just to get a particular question deleted would take a maniac. Besides, the OP would naturally respond by creating a meta thread calling out a fake, and it would not take long for meta users to set the record straight. –  user31373 Sep 29 '12 at 14:50
    
@LVK: Yes, while it is also possible to do what mt_4 describes, maybe the incentive is a lot lower (for the average non-maniac). Then again, the average non-maniac won't be too interested in IP data either - but you know what kind of people are "out there". –  Hagen von Eitzen Sep 29 '12 at 14:58
7  
@LVK it certainly is far fetched (though people sometimes go to surprising lengths to harass others). I do think it illustrates a point: mods would have to decide which contests/homework to police, whose emails are to be taken seriously and so on. I know some people think maths contests are serious business but I don't think we have to buy into this. Personally I wouldn't answer something that I knew to be part of an open contest, just like I wouldn't answer homework where the OP hadn't made an effort. But I don't think we should ask the mods to be content police. –  mt_ Sep 29 '12 at 14:58
4  
@mt_: Right. I'd encourage everyone to leave a hint that there is an ongoing contest (until which date?) to allow others the option to refrain from unwillingly help cheaters. –  Hagen von Eitzen Sep 29 '12 at 15:01
    
@MadScientist Alas, the SE privacy policy might not be as strict as we think. For example the recent mass deletion of purported contest questions also included the same mod revealing to an external entity the fact that some users were posting from the same IPs. By cross-matching, this allowed the external entity to uniquely identify the user. One SE spokesperson said he thinks this is allowed by the SE privacy policy. I think we should convince them to reconsider. This could easily be abused by oppressive regimes etc with just a little social engineering, e.g. as Hagen suggests. Thoughts? –  Bill Dubuque Dec 12 '12 at 4:24
6  
@BillDubuque, your comment is misleading. Our policy is stricter than many sites, and the implication that the information we provided allowed someone to ID a user they otherwise couldn't is unsupported. In both contests, the cheaters had been identified; the question was more around whether they were two bad actors or one. That said, in the interest of candor, while Qiaochu's actions don't violate the policy, I think I agree with your main point that mods should not share any info about IPs, real names, or emails with any third party, whether or not they happen to be an oppressive regime. –  Jaydles Dec 12 '12 at 19:27
1  
@Jaydles Not true. A moderator supplied information on relationships between account names and IP addresses that aided an external entity to personally identify an MSE user (or so they think, often these inferences are incorrect). In my opinion this is a gross violation of privacy. Moreover, if SE is going to condone such rogue moderation decisions, then it is the (vague) SE privacy policy itself that is misleading. –  Bill Dubuque Dec 12 '12 at 19:32
add comment

Done. Some clarifications for everyone else:

  • As Mad Scientist says in the comments, the SE privacy policy does not allow us to release anyone's IP addresses. This is not under our control.
  • I strongly disagree with mixedmath's answer in the linked thread. When it is brought to my attention that a question is from an ongoing contest, my policy is to delete it until the contest is over. (Deleting a question is reversible, but openly posting a solution to an ongoing contest problem is not.)
share
14  
I don't think it is our business to enforce rules of some external entity. –  Bill Dubuque Sep 27 '12 at 20:28
6  
Moreover, moderators should not execute unilatateral deletions on such contentious matters, esp. since such deletions cannot be reversed by the community. –  Bill Dubuque Sep 27 '12 at 20:35
16  
But if many users support such deletions, we would want a moderator to be able to do it in a timely fashion without having to wait to gather a couple of other users with very many points. I think that often "unilateral" action is better, with the understanding that we can voice concerns and request policy change on meta. –  Jonas Meyer Sep 27 '12 at 20:40
16  
@Bill: I strongly disagree. The whole point of any kind of action in a situation like this is to act before other people participating in the contest see the question. Waiting to make a decision until there is community consensus is practically equivalent to deciding not to delete the question. –  Qiaochu Yuan Sep 27 '12 at 20:43
10  
@Bill: I am not trying to teach anyone morals. –  Qiaochu Yuan Sep 27 '12 at 20:50
13  
If a question has no answers, then temporary locking is sufficient. If a question has answers, then I see two possible temporary actions: 1. delete the question; 2. delete the answers and lock the question. These actions should be reversed as soon as the contest/homework is over/due. Temporarily locking the question has the benefit of letting others know that the question is part of a contest/homework. –  robjohn Sep 27 '12 at 21:13
9  
@Robjohn It certainly is your prerogative to decide not to answer it. But it is not your prerogative to make that decision for others. –  Bill Dubuque Sep 27 '12 at 21:36
12  
@BillDubuque: Indeed, the charter of this site is to teach mathematics. However, if we immediately answer questions that we know are on contests/homework, we do harm: harm to the others in the contest; harm to the student who does not learn from their homework; harm to the people who rely on someone, thinking they know things they don't. All decent questions posted on this site deserve an answer, but I see little harm in delaying that answer to an appropriate time. –  robjohn Sep 27 '12 at 22:27
10  
@Robjohn The situation is highly analogous to homework questions The most reasonable policy is to let every user decide for themselves if/when/how they will answer based on whatever information is available about external matters. We should not cripple our site based on matters having nothing to do with us. It is the job of contest designers to take into account the large amount of knowledge easily accessible to participants. That is not our job. –  Bill Dubuque Sep 27 '12 at 22:36
11  
Would this be acceptable to everyone/anyone? When a moderator becomes aware that a contest problem has been posted, the moderator edits into the question something to the effect of "This is a problem in the such-and-such contest. The contest closes on such-and-such a date. Please take this into consideration when thinking about posting a response." –  Gerry Myerson Sep 28 '12 at 0:29
9  
I completely agree with Qiaochu and robjohn. And I do not agree that deleting/stalling contest questions would "cripple our site" as Bill says. In fact, I would like to think of MSE as beneficial to the mathematical community as well as individual students learning math - and to allow contest questions to be answered here harms the integrity of those contests. In addition, I am going to edit my answer that Qiaochu has referred to, indicating that this thread is the most recent thread about the idea of answering ongoing contest questions. –  mixedmath Sep 28 '12 at 1:03
10  
@mixedmath That's a very slippery slope. If you start allowing questions to be deleted because they are similar to questions elsewhere, then that opens the door to many analogous requests, e.g. temporarily delete older questions because they are current homework, take-home exams, etc. This is not our problem. Rather, it's a problem with folks using antiquated methods for tests, contests, etc. - methods that are a poor fit to the current information age. Any problems they encounter should be fixed at the source - not kludged here. –  Bill Dubuque Sep 28 '12 at 1:15
11  
@Bill It may be a problem for MSE, specifically for its reputation (the one not measured in points). If the site comes to be perceived as a cheaters' paradise a la wedoyourhomework.com, MSE will have a hard time attracting new expertise that it sorely needs. Imagine all 12 Putnam problems posted and solved here on the morning of Dec 1 - what will be the reaction of math community? –  user31373 Sep 28 '12 at 2:11
6  
There are clearly costs and benefits to enforcing any policy. I think temporary locking as advocated by Robjohn is a great compromise. I don't like slippery slope arguments in a world that is full of shades of grey. –  Michael Greinecker Sep 28 '12 at 10:48
11  
@Bill: I don't know what about you, but seeing how I hope to have a future in the academy, I would very much like to avoid having a community I greatly participate in flagged as "cheaters heaven". I am fairly certain that I am not the only one. –  Asaf Karagila Sep 29 '12 at 14:42
show 23 more comments

This is the complementary answer to my other poll question.

VOTE THIS ANSWER UP TO OPPOSE the moderators rapidly deleting or closing questions from ongoing contests, without waiting to seek consensus from the broader community.

(This answer is CW.)

share
6  
I strongly oppose deleting or closing questions from ongoing contests. This is not about moderators vs the community, it's because I am opposed to the idea of giving contest organizers a monopoly on a mathematical question. I am disgusted that the Math.SE community seriously entertains this radical departure from scientific and academic traditions that do not recognize a monopoly on a scientific idea. –  Gilles Sep 30 '12 at 21:27
6  
@Gilles Contest problems are not research problems. Contest problems are specially designed artificial questions meant to be solvable in the timeframe of a competition. Moreover, contests end and then the problems are available to the world. –  David Speyer Sep 30 '12 at 21:34
2  
Why should there be a difference between research problems and problems with a known solution? We had a similar issue on Computer Science with questions from a textbook where the author asked not to publish solutions: would you suppress such questions as long as the textbook is in print? If the question is not original (most aren't), what gives the contest author any right to it? –  Gilles Sep 30 '12 at 21:40
5  
Umm, I don't know what contests you've worked on, but every contest I've been involved with took a great deal of pride in producing original problems. (I've written for the HMMT and submitted problems to the USAMO and Putnam.) –  David Speyer Sep 30 '12 at 21:55
1  
The issue of old chestnuts, like the fly going back and forth between two trains, has more grey areas. On the one hand, it would be silly to say that these questions should never answered online because someone might put them in a contest again. On the other hand, if I saw someone had actually pasted the specific phrasing of it from a math contest, I would post a comment urging others not to answer, because in that case answering would be helping a cheater. –  David Speyer Sep 30 '12 at 22:00
2  
"Specific phrasing" is copyrighted, which again suggests that the legal standard may be the one to use here rather than holding the site hostage to math competitions, homework, take-home exams and other such situations. There are hundreds of contests and thousands of the other categories at any moment. –  zyx Sep 30 '12 at 22:02
1  
Actually, I dispute that there are hundreds of contests which take place in an unproctored environment. I know of two: USAMTS and OMO . While proctoring is of course not perfect (especially in cases where local schools proctor nationwide contests, so there is a risk of proctors collaborating with cheaters) I imagine that simply being aware of these contests would cover most of the issues. Take home exams are a much harder problem. –  David Speyer Sep 30 '12 at 22:09
2  
@David: as well as USAMTS and OMO, there are numerous admissions problem sets for summer programs. –  D. Savitt Sep 30 '12 at 23:04
1  
@D.Savitt Good point. That might be a harder case to deal with since I know that PROMYS, at least, reuses the same questions. That starts to run into Gilles' concern about whether a problem can be eternally off limits. –  David Speyer Sep 30 '12 at 23:33
3  
Canada/USA Mathcamp learned long ago that we shouldn't reuse questions, partly because we can't keep answers off the web indefinitely, partly because we think the problems are interesting and we do want to encourage discussion of them in the long run. As a result we go to a lot of effort every year to write engaging problems that test what we want to test. As a part of the mathematical community, we certainly appreciate the efforts of the broader community to keep things on the up-and-up in our admissions process. –  D. Savitt Oct 1 '12 at 0:02
3  
@D.Savitt Good to know! Does Mathcamp put the questions for the current year online, so that we users here can spot them if they are asked? –  David Speyer Oct 1 '12 at 0:04
3  
Absolutely! See mathcamp.org/prospectiveapplicants/quiz/index.php for last year's quiz (the committee to write next year's quiz just started working in the last couple of weeks). –  D. Savitt Oct 1 '12 at 0:06
2  
There are dozens, and probably a hundred or two when taken internationally, of journals with problem columns, plus problem-of-the-week competitions and web sites (the IBM one was fairly prestigious). Plus national IMO selection and training programs that include problems solved by correspondence. Plus some qualifying exams in graduate school. I could continue the list, but clearly the number is a lot higher than two or twenty. –  zyx Oct 1 '12 at 14:12
add comment

You must log in to answer this question.