The following thirteen(!) users:


(and, I believe, a couple more besides) have the following characteristics in common:

  • They've all joined in the past few days.
  • Each has asked exactly one question
  • Their questions are closely related to each other
  • Their writing styles are very similar
  • After posting their questions, they never return to engage with commenters, or to acknowledge answers

I think it is not too much of a leap to conjecture that these are all the same person.

As far as I know, there's no actual rule against having fifteen different identities (or for that matter, a hundred, as seems likely if this keeps up at this rate for a few more weeks), and no actual rule against asking questions and then systematically ignoring responses, but I still find it troubling in ways that might or might not be entirely rational on my part. Do others share my discomfort?

Edit (YCor June 3, 2019): since the question was posted, here are a number of similar posts:

user138661 (account closed since then: not the original post name)
Gimme An Answer
La fille
Name (not the same as previous one)

(there are a few more, for which the similarity is less obvious). In addition, this June 3, 4 of these posts have been bountied by other users: cardinal, Cut the wood, kartop_man (this one and this one, and now for the time being self-deleted: https://mathoverflow.net/users/141715/kartop-man).

Edit: I'll add more as I see them. Please note that these accounts have a tendency to comment on, edit, and answer each other's questions, or plan on offering each other bounties. The evidence I see points to one human in particular behind all this, and the situation is similar to that of another user and his ring of accounts a few years ago that some of you may recall, and who is now permanently banned from SE.

New Asura Path, possibly temporary: https://mathoverflow.net/users/141752/asura-path?tab=profile
New https://mathoverflow.net/users/141770/phd
https://mathoverflow.net/users/141790/numberian (now 141790, 141792, 141793, 141794, 141795)

(YCor 2019/July/12) A last fact, of one higher level in misbehavior (if confirmed), is the use of an username that is a seemingly real name (Denis Chaperon de Lauzières), which has been a real user of MO (real accounts here and here). Two suspicious posts under 1 user with this name were posted here and here (user now removed and both question deleted by community), and a comment-like answer here (same name, other account - now converted to a comment).

  • 10
    $\begingroup$ I have over ten accounts on the Stack Exchange network myself. (This account is the one that I use most frequently as of late.) While I understand your concern, I think the content of the questions should be the issue, not the manner in which they were asked. Also, as user names can be changed, you might list user IDs instead. Gerhard "No, I'm The Other One" Paseman, 2019.05.01. $\endgroup$ May 1, 2019 at 22:55
  • 24
    $\begingroup$ @GerhardPaseman : There are some relevant differences, though. For one thing, I'm not counting "accounts on the Stack Exchange network"; I'm counting accounts here on MO. For another, you have multiple accounts under names that are recognizably yours, which seems a little less suspicious. For another , your multiple accounts did not (I believe) all spring forth full-blown nearly simultaneously. And for yet another, you have no substantial history of hit-and-runs, where you ask a question, a commenter responds, and you're never heard from again. $\endgroup$ May 1, 2019 at 23:16
  • 22
    $\begingroup$ Gerhard: Well, one issue you're overlooking is that I might not want to put much effort into this person's questions in the future, given the likelihood that my responses will be ignored. If the username keeps changing, it's a little harder to identify the questions I prefer to skip by. $\endgroup$ May 2, 2019 at 0:26
  • 21
    $\begingroup$ Another explanation might be: MO was mentioned by the instructor in some math class somewhere, and as a result 10 of the students decided to try it. $\endgroup$ May 2, 2019 at 10:37
  • 12
    $\begingroup$ I would like to say in defense of these questions that several of them are interesting, such as mathoverflow.net/questions/330337 , mathoverflow.net/questions/330178 , mathoverflow.net/questions/330168 , mathoverflow.net/questions/330159 , and all of them are at least reasonable. This is different from other times when we've had a flood of what are clearly low level homework questions. $\endgroup$ May 2, 2019 at 15:09
  • 9
    $\begingroup$ @StevenLandsburg I don't think either of us are disputing each other; I just want to make sure people who are looking at this thread have a good understanding of the situation. $\endgroup$ May 2, 2019 at 15:12
  • 6
    $\begingroup$ @MartinRubey I would love to see the code if all of these were written by a program. It got at least some of the semantics right. $\endgroup$
    – user138661
    May 4, 2019 at 11:13
  • 9
    $\begingroup$ @schematic_boi: I strongly believe that you are the one behind all these usernames. $\endgroup$
    – Alex M.
    May 21, 2019 at 17:08
  • 16
    $\begingroup$ @cardinal This is probably OP, but I cannot imagine posting a question on MO without spending at least a day thinking about it to see if it is really nontrivial. And even so, I'd probably spend at least 20 minutes writing the question, to make sure I am satisfied with its formulation. In this light, you'll see why your "someone could produce a question every ten minutes" strikes me as ridiculous at best. $\endgroup$ Jun 4, 2019 at 15:24
  • 9
    $\begingroup$ @DenisNardin this is backed up by the FAQ: Do your homework Before asking your question, try to solve it. Search Google, Wikipedia, and nLab, check any references you can think of, and try to figure the problem out yourself (maybe even sleep on it). $\endgroup$
    – David Roberts Mod
    Jun 5, 2019 at 6:18
  • 15
    $\begingroup$ @StevenLandsburg What I find vexing is that to my non-AG eye, some of these questions look reasonable; but I really dislike the disingenuous behaviour of different user accounts interacting as if they are different people, even if this does not involve voting. I also find the creation of "one-shot" or "burner" unregistered accounts solely for the purpose of a high-volume of questions verging on obnoxious spamming $\endgroup$
    – Yemon Choi
    Jun 6, 2019 at 12:56
  • 9
    $\begingroup$ Because that account doesn't fit the pattern we're seeing. I don't see why it should make any difference to you, if you claim you have nothing to do with these other accounts. $\endgroup$
    – Todd Trimble Mod
    Jun 10, 2019 at 22:38
  • 14
    $\begingroup$ @StevenLandsburg Yeah. I've now got some Community Management looking at the case, and we're pretty much in agreement with what's going on. Meanwhile, a very large number of accounts have been deleted. $\endgroup$
    – Todd Trimble Mod
    Jun 14, 2019 at 2:17
  • 9
    $\begingroup$ @ToddTrimble: once again, I'm very glad you're on this. $\endgroup$ Jun 14, 2019 at 2:21
  • 7
    $\begingroup$ at the potential cost of jeopardizing his immediate future, is something I’m having a very hard time understanding. Everybody makes mistakes: what’s important is to draw a lesson from them and turn them into, precisely, “a positive drive for change”. I’m hoping you’ll take the following suggestion in a constructive way, and act on it, after taking some time for yourself to reach this conclusion on your own: leave a short message to the Mathoverflow community, down here, apologizing for causing so much confusion, and unnecessarily wasting a number of people’s time. $\endgroup$
    – user135665
    Jun 24, 2019 at 0:41

4 Answers 4


With regard to Gerhard's response -- which I agree with -- I see a number of things problematic in some of the behavior taking place in the present circumstances.

I have no doubt that behind these numerous accounts is only a very small group of humans, and there's a very good chance it's just one. In other words, I am finding very clear evidence, using the moderator tools available to me, of sockpuppetry taking place. As you all know, sockpuppet upvoting is a significant violation of site (and SE) policy, but sockpuppetry is obnoxious for other reasons as well. In the present case, there have been numerous instances of the sockpuppets' answering each other's questions, editing each other's posts, and leaving comments on each other's posts. Even in cases where no sockpuppet upvoting is taking place, an appearance is trying to be created of great community activity and interest surrounding the posts in question. This type of attention-drawing is based on deceit, is thoroughly obnoxious, and is inherently problematic.

I think the behavior promised in this user profile https://mathoverflow.net/users/141500/human-path?tab=profile also qualifies as problematic:

The Six Paths of Pain are controlled by Nagato. The Six Paths of Pain do not answer, upvote, comment on each other's posts. The Six Paths of Pain award bounties on each other's posts.

Judging from this post, it seems that Community Management agrees this behavior is not cool: Who or what is this SO user?

Still other problematic aspects include opening new accounts in response to other accounts being closed. In one case relevant here, an account named "mika" was (re)created after a same-named account was deleted, with the user openly saying so. Deletion can mean one of several things. One is that Community Management or site moderators have decided that the user is no longer welcome, in which case the user opening a new account is explicitly in violation of community moderation. Another is that the user is self-deleting the account voluntarily. But in this case as well, coming back again in a new form flouts the norms of this site, especially by enabling the user thereby to evade community moderation and policing. This is akin to question repetition, as when a user deletes his post and then asks again.

Still another problem -- more of a judgment call but nonetheless real -- is trollish behavior (sealioning, etc.). In my opinion, and in the opinion of other community members, there have been a few instances of that in the current circumstances, both at meta and at main.

In such cases, when the behavior becomes problematic and repeatedly so, we may have to clamp down a little harder. (Many of you will remember one or a very small band of users engaged in shenanigans a few years ago, with the main perpetrator now permanently banned from SE.) I don't have a problem per se with a user controlling a number of accounts -- as long as site policies and guidelines are not being flouted and deceit is not taking place. Otherwise I have a big problem with that.

I guess I'll begin documenting here suspicious events as I see them. For those users who seem puzzled about my linking certain posts on MO Main to this discussion, I repeat: forensic analysis strongly suggests to me that all of the many accounts listed above are associated with each other, and I think the community should be aware of events that are occurring within this circle.

For those with 10k+ reputation: the (unregistered) user "Curious girl" asks a spam question https://mathoverflow.net/questions/333692/big-topos-guys-why which makes a crude joke about genital size.

User "cardinal" https://mathoverflow.net/users/74900/cardinal has the following recent reputation events https://mathoverflow.net/users/74900/cardinal?tab=reputation: yesterday there are 5 upvotes in the same minute 20:14, and 4 upvotes in the same minute 16:26. Today (June 11) there is a slew of upvotes all within a short span of time.

A number of accounts in this circle that are offering bounties follow a characteristic pattern where coincident with offering bounties, the account is scheduled for deletion. This will continue until past the time of deletion, and then the deletion will be rescinded.

A new version of Asura Path became a member today https://mathoverflow.net/users/141752/asura-path?tab=profile, after a same-named account was self-deleted. This self-deletion plus resurrection may be an attempt to circumvent ordinary moderation. Here is a list of posts by the old Asura Path: https://mathoverflow.net/admin/posts-by-deleted-user/141498

  • 13
    $\begingroup$ Thank you for being on top of this. $\endgroup$ Jun 3, 2019 at 22:15
  • 13
    $\begingroup$ Seconded. Thank you not just for acting, but for posting this answer to clarify for those who might not have seen the recent activity $\endgroup$
    – Yemon Choi
    Jun 3, 2019 at 22:56
  • 4
    $\begingroup$ If I understand correctly, these are all unregistered accounts. These can't vote. $\endgroup$
    – Asaf Karagila Mod
    Jun 5, 2019 at 6:50
  • 4
    $\begingroup$ @AsafKaragila It's a mix of registered and unregistered. Moreover, unregistered accounts can engage in shenanigans, as I explained in my answer. $\endgroup$
    – Todd Trimble Mod
    Jun 5, 2019 at 10:15
  • 4
    $\begingroup$ @YemonChoi Thanks for the suggestion! $\endgroup$
    – Todd Trimble Mod
    Jun 6, 2019 at 13:00
  • 15
    $\begingroup$ I want to publicly voice my support for a policy of forcibly deleting these accounts as they arise, if the moderators find that action warranted. $\endgroup$ Jun 11, 2019 at 23:23
  • 4
    $\begingroup$ @StevenLandsburg Thanks for your support. Such an option is on the table. $\endgroup$
    – Todd Trimble Mod
    Jun 12, 2019 at 10:37
  • 4
    $\begingroup$ One of the users mentioned has recently posted a 100 rep bounty on a question and I'm confused as to how they've got the reputation for that. According to their profile it comes from the association bonus, but there are no signs of their activity on other SE sites either. $\endgroup$
    – Wojowu
    Jun 12, 2019 at 14:43
  • 5
    $\begingroup$ @Wojowu This mystery is explained in the comment thread of meta.mathoverflow.net/questions/4261 . The SE software actively assists abusers by making it possible to hide communities from your profile. $\endgroup$ Jun 13, 2019 at 8:00
  • 6
    $\begingroup$ I find it enormously funny that the now deleted account of user141498, which is one of the accounts involved in this scheme, joins the conversation and shares some common sense wisdom on the subject with us. $\endgroup$
    – Alex M.
    Jun 13, 2019 at 18:10
  • 4
    $\begingroup$ @ToddTrimble You are probably aware of it, but I will explicitly mention that the link you've included - mathoverflow.net/admin/posts-by-deleted-user/141498 - is not visible by regular (non-diamond) users. Regular users can get list of posts from a deleted users using SEDE, as discussed before on this meta, this does not include deleted posts. There are some stats about deleted users that might be interesting in connection with this, I have mentioned some of them in chat. $\endgroup$ Jul 1, 2019 at 5:13
  • 4
    $\begingroup$ mathoverflow.net/users/142965/gentbent appears to have acquired an association bonus by responding to trivial questions from augur.stackexchange.com/users/1822/maoboa on the Augur stackexchange, getting one upvote each time. $\endgroup$ Jul 12, 2019 at 15:37
  • 4
    $\begingroup$ @RobertFurber yes, and also on a bunch of other sites, making them look reasonably engaged. $\endgroup$
    – David Roberts Mod
    Jul 13, 2019 at 0:48
  • 5
    $\begingroup$ @GetOffTheInternet One likely scenario is that the 'real' Aknazar, having been outed, claims that someone he knows has hacked his account and is maliciously using it to frame him. Then the same Aknazar creates this entity who has it in for him, thereby justifying the claim. $\endgroup$
    – Todd Trimble Mod
    Jul 15, 2019 at 16:52
  • 5
    $\begingroup$ So far I don't think the last two meet the threshold. $\endgroup$
    – Todd Trimble Mod
    Aug 3, 2019 at 22:43

I believe that I can give a partial answer to your question. Looking at some of the profiles that you have listed, they seem to be all unregistered users. I used to do something similar (but with only 4-5 names) in 2012, at my beginnings on MO: not in bad faith, but because I was viewing my asking questions on MO as only a temporary thing. Plus, I was relying on the cookies mechanism to remember me between browsing sessions. Alas, somehow those cookies used to disappear after some time, which forced me to create a new unregistered name every such time. When I decided that MO is a useful community where I'd like to stay, I merged all of my unregistered usernames into the current, definitive one.

The solution to your question would be to tell this person in a comment to do the same: register and then ask the SE staff to merge all these usernames into a single account.

The longer-term solution has been previously discussed, but has been decided against: that unregistered questions should no longer be allowed, following the example of Math.SE. (Back then the issue was settled in the negative based on the consideration of solely the quality of the questions.)

  • 1
    $\begingroup$ Alex, that was prior to the merge into the SE network. This made merging unregistered accounts a bit slower, a bit more cumbersome, and out of the hands of local moderators. $\endgroup$
    – Asaf Karagila Mod
    May 11, 2019 at 23:49
  • 2
    $\begingroup$ You are right that when I posted the question about registering I should have stressed this aspect too - if a question is posted by an unregistered user and they lose the account, it makes more difficult for them to follow the question, add comments, etc. $\endgroup$ May 12, 2019 at 8:58

Do(es)n’t said user(s) simply exploit the loophole once detected by @MattF. (newly created accounts can spend the association bonus on bounties), and wouldn’t that part be solved by following his then-suggestion (make the “set bounties” privilege higher than 101)?

  • 1
    $\begingroup$ The new incarnation of Asura Path got an association bonus both here (which they spent almost immediately on a bounty) and on math.stackexchange, although I can't see why they qualified on either site. $\endgroup$ Jun 12, 2019 at 10:28
  • $\begingroup$ Also the second incarnation of “phd” got an association bonus and spent it almost immediately on a bounty, despite apparently being a member of no other SE site to qualify for the bonus. $\endgroup$ Jun 12, 2019 at 13:14
  • 1
    $\begingroup$ It's a bit more difficult how much reputation a deleted user earned on MO - that could be used as an estimate whether some of the reputation came from association bonus. I have tried to created some SEDE queries. For some users we'll have to wait for the next database update - for two users (138661 and 140765 - I saw that they earned enough reputation on MO. $\endgroup$ Jun 12, 2019 at 13:39
  • 4
    $\begingroup$ As transpired on meta.mathoverflow.net/questions/4261 , it is in fact possible to hide SE communities from someone’s profile. As far as can see, this misfeature is just a recipe for abuse. $\endgroup$ Jun 13, 2019 at 7:57

I think it is reasonable to be concerned, if only for the fact that we do not know what the ramifications of this behaviour are.

However, I think it is better to describe the behaviour and call out what might be problematic about the behaviour. Right now, it seems we are getting OK to good quality material for the forum, and we have the ability to improve upon and respond to this material. There is no suggestion of copyright, homework or exam flaunting, or other misbehaviour.

Bringing the situation to the attention of the community is good, but I don't see the quality or quantity a reason for concern.

Gerhard "Having One Name Is Boring" Paseman, 2019.05.02.

  • 18
    $\begingroup$ Gerhard, you have so many middle names, that it would take a small book to write them all. Asaf "This is not my real middle name" Karagila, date of comment is written after my name is written again by the system. $\endgroup$
    – Asaf Karagila Mod
    May 3, 2019 at 9:32
  • 17
    $\begingroup$ I look forward to the publication of The Many Middle Names of Gerhard "Whose Idea Was This, Anyway?" Paseman. $\endgroup$ May 3, 2019 at 10:16
  • 6
    $\begingroup$ Volume II-M is available here: meta.mathoverflow.net/users/3402/gerhard-paseman?tab=activity . Gerhard "Nine Other Volumes In Preparation" Paseman, 2019.05.03. $\endgroup$ May 3, 2019 at 14:58
  • 25
    $\begingroup$ How about the following: Gerhard "Recursion never goes wrong Gerhard "Recursion never goes wrong Gerhard "Recursion never goes wrong Gerhard "Recursion never goes wrong Gerhard "Recursion never goes wrong Gerhard "Recursion never goes wrong Gerhard "Recursion never goes wrong Gerhard "Recursion never goes wrong Gerhard "Recursion never goes wrong Gerhard "Recursion never goes wrong Gerhard "Recursion never goes wrong Gerhard "Recursion never goes wrong Gerhard "Recursion never goes wrong Gerhard "Recursion never goes wrong Gerhard "Recursion never goes wrong Gerhard "Recursion never goes wrong $\endgroup$
    – user138661
    May 4, 2019 at 13:16
  • 12
    $\begingroup$ Q: “What does the B. stand for in ‘Benoît B. Mandelbrot’?” — A: “Benoît B. Mandelbrot” 😉 $\endgroup$
    – Gro-Tsen
    May 31, 2019 at 15:59

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .