Not signed in (Sign In)

Vanilla 1.1.9 is a product of Lussumo. More Information: Documentation, Community Support.

    • CommentAuthorquid
    • CommentTimeOct 17th 2011

    I was recently made aware of the fact that some (or at least one) person(s) are (is) concerned about the fact that anonymous users vote to close and reopen questions (for source and context, please see below).

    Out of general considerations, as well as being (one of) the reason(s) for this concern, I would like to ask the community for opinions on this matter, and in particular I would like to ask those for whom this should be a concern to voice it, if possible, including a reason for it.

    Some remarks:

    1. This is not intended as a general debate on anonymous usage of MO (this has happened in the past, and in case somebody should like to restart it, s/he can count on my participation, but please do not start it in this thread). Only whether the act of voting is a specific concern is the subject of this thread.

    2. How frequent are anonymous users that vote: this is a bit hard to tell as it depends on ones precise definition of 'anonymous', and which one is relevant depends on the precise reason for the concern. In any case, under most typical definitions, a few percent of the entire voting user pool (although there are certain ramifications to this question that would increase this percentage to a more significant fraction, though perhaps I should not digress).

    3. This debate is not entirely abstract, since at least I am willing (in certain ways) to react to concerns that should be raised. However, which line of action would be useful to address the concerns depends on their precise nature, the reason for the concern. Thus, please, supply it.

    Source and context:

    The concern was raised by Gil Kalai in a discussion with me in the meta thread of the "Common false believes" question (towards the end). I did not quote him here directly, since this would more or less make it necessary that I also rigth away 'reply', and this would make this starting post a bit long and potentially push the discussion in a certain direction. [@Gil Kalai: could you please reraise your concern here, even if only by coping it, thank you.] That he is only/specifically concerned about the voting, however, informed the precise phrasing of this post (I hope I am not commiting a misinterpretation, yet it seems very clear).

    Summary and repetition:

    This thread is for opinions and concerns regarding the fact that anonymous users vote to close and reopen questions , not about anonymous users in general.

    Personal note: I intend to participate in the discussion, but likely not initially (it depends a bit; I will reply to questions that diretly involve/address me and supply factual information (to the extent I am able to) in case the need should arise without delay, except for the inevitable one).

    • CommentAuthorgrp
    • CommentTimeOct 17th 2011

    Question for entity with the handle "quid": Do the moderators have a good idea who you are? I don't mean just an email address; could they make a statement like "I know who quid is; she works in the same department as Gerry Myerson." , or something similar?

    This question will go towards accountability, but I prefer an answer with some detail, without necessarily compromising your anonymity more than you already have.

    Gerhard "Ask Me About System Design" Paseman, 2011.10.17

    • CommentAuthorquid
    • CommentTimeOct 17th 2011

    @grp: At least with some effort (not sure how large) I assume they could know who I am; and that it perhaps is an effort is more coincidence than intent. They also could ask me, and I would not hesitate to supply the information to them. There is also one MO user with whom I coresonponded specifically on MO related matters via an email address that makes my identity completely transparent.

    However, I am not at all sure how this is relevant to voting or what type of accountability you mean. Voting is not something which is easily abusable or to be more precise even if it is is abused it causes no great disruption to the workings of the site. As long as not four others (or a moderator) agree with me, it is essentially irrelevant how I vote, and same for anybody else (except for moderators). Editing privileges seem much more disruptive when abused, and one can act unilaterally. Even just posting many questions, or arbitrary downvoting or flagging seems more disruptive. And even if 'a-gang-of-five-anonymous-voters' should form (but at least for the most part I assume we/they do not know each other either) they could once close a specific question, and then it could be reopened and then it is over (as one can vote only once on each question).


    I can say, I don't know who quid is, (I don't even have an email address) but I haven't tried very hard to find out.

    • CommentAuthorgrp
    • CommentTimeOct 17th 2011

    From my point of view, it is enough that someone who helps moderate the site can track down a vote, tie it to an individual (more precisely, a packet of information that is associated with said individual), and use it to determine propriety, motive, or whatever else is needed to ensure the integrity of the voting process.

    Here is a scenario to place force on some of the issues. A question appears, and then gets closed. The list of users who closed it are Gerhard Paseman, Gerhard Paseman, Gerhard Paseman, Gerhard Paseman, and Gerhard Paseman. (The reason for closing is not pertinent, but lets say it was closed as a duplicate question, for sake of whimsy.) Is this a valid closure, or an abuse of the voting system?

    The correct answer is that not enough information is given. Although I have more than 5 accounts on MathOverflow that I maintain, I use only one of them for voting. However, I could have let four other people use these accounts with my permission and knowledge, and that the five of us agree on the handling of this question.

    While the validity of the closure could be discussed, one thing of which I am certain is that I would be contacted by the moderators to make sure of such validity. If I had no arrangement with them (no accountability) of how the accounts are being used, I would likely see one or more of the accounts being suspended or yanked.

    At present, I am accountable for the activity on all the accounts I have, and at least one of the moderators has my personal contact information, so that they have more than usual access to a user to resolve such matters, if such is ever needed.

    Indeed, one difficulty with a nom de plume would be to get me to call you by any other name.

    Gerhard "Ask Me About System Design" Paseman, 2011.10.17

    • CommentAuthorabatkai
    • CommentTimeOct 17th 2011
    I have to confess, I do not understand the problem. You need at least 3000 rep to vote. If you have 5 accounts, each of them with 3000 rep, then you must have contributed something to the site. The assumption is, it is not possible to be a troll with such contributions...
    Though I am personally not anonymous, I completely sympathize with quid.

    I think that there's no good reason for people to have multiple accounts and vote to close a question twice. I think such behavior counts as mixing the accounts (like voting on your own answers) and is grounds for bannination. I don't think there's much reason to worry about it beyond that. Unless there's specific closing behavior that's worrisome, I don't see why pseudonymity itself is the problem.

    (That said, I still believe that people should use names that make it clear who they are. I think it's good for MO to have less pseudonymity.)


    I cannot see an argument against anonymous users voting to close/reopen. Though I think it's worthwhile to cultivate a culture of attaching ones professional identity to ones username, there is no reason to prevent any particular person from participating pseudonymously. Voting with multiple accounts (never mind voting to close with multiple accounts) is a no-no. But 3000 reputation is hard to get and easy for a moderator to take away (via suspension), so it's hard to imagine ever having a serious problem.

    To anybody who is concerned with users like quid voting to close/reopen, I'd like an answer to the following question. What if instead of the handle "quid", s/he used the handle "Victor Chen" (a name I just made up, but is surely "owned" by a real person)? Supposing Victor was sufficiently invested in MO to post 100 answers, generating 4000 reputation, on what grounds could you argue against this user voting to close/reopen questions? What amount of identifying information would be necessary to make you happy?

    To put it another way, once you get to 3000 reputation, why should your MO account be any worse a mechanism of identification than your driver's license? From a moderator's point of view, it's better, since you can't "fake" your MO account.

    [Edit: btw, I don't know quid's real life identity]

    • CommentAuthorMariano
    • CommentTimeOct 17th 2011

    On the other hand, I don't think there is any reason for people to have multiple accounts...

    • CommentAuthortheojf
    • CommentTimeOct 17th 2011

    If I had my druthers, every MO account would have a full name attached to it, and a link to an academic website. I like to look up arXiv publication lists, or remember what institution someone is at so that if I visit I can look up the people I've met online, or... But that's a different discussion topic.

    To quid: You are a very regular contributor to MathOverflow. I believe that you are the only user who uses the handle "quid" (whereas I am not the only user who goes by "Theo"!). I have a sense of your mathematical opinions and meta.MO opinions, and respect them. And most importantly, I know that you participate in the community. So from my point of view, when you vote to close a question, or comment on meta, or whatever, you are doing so under your own name, and there's nothing pseudonymous about it.

    Conversely, I really don't like it when "unknown (google)" leaves comments or votes or .... If there were only one unknown (google), and s/he were a known quantity, I would care less. But I can remember names much better than non-names, and I certainly can't keep straight users with strange handles.


    Occasionally people want to ask a question anonymously. This means that occasional second accounts are reasonable. I don't see any reason for them to get to 3000 rep though.

    • CommentAuthorYemon Choi
    • CommentTimeOct 17th 2011

    Just passing through while silently screaming about the wave of overdue work that's fallen on me, but: regarding Theo's 2nd paragraph, I'd say that quid is being pseudonymous but not anonymous .


    I have no issue with anonymous users voting-to-close. Indeed, for me then the more important thing is that the users have a meta account so that when there is a dispute it can be discussed. Since this is (currently) our forum for such discussions, then having an account here is more important in my eyes than being identifiable as a real person.

    That said, I agree with Theo: purely from my own use of the site, I find things much easier if I have a name to link things to. I've argued this elsewhere so shan't say more on this here (search for "Bugs Bunny" and "Bill Johnson" in the discussions). My desire is that there be a culture that encourages real names (and links to academic websites if available) but which does not actively punish anonymous/pseudonymous.

    (That said, I'm curious as to quid's reasons for being anonymous. Purely from interest since I find it hard to believe that a user has gotten to that level of involvement without thinking seriously about whether or not to remain anonymous so my curiosity is directed at better understanding why a user might want to remain anonymous.)

    • CommentAuthorgilkalai
    • CommentTimeOct 18th 2011 edited
    Since there is some confusion what anonymity is: I suggest to regard in this discussion anonymous user as one whose identity is unknown to others. If the user's name is just his first name or a nickname but there is a link to his homepage then he is not anonymous for the sake of this discussion. If a person has a full name that we cannot associate to any "real" person then he is anonymous.

    Regarding anonymity in general. There was a long discussion about it. My own view is that participants should have the privilege to post questions and to answer questions anonymously, and I see a few reasons why people will choose not to identify themselves when asking or answering questions. Of course, the quality of a mathematical answer and question can be judged on its own. Since casting votes is done anonymously for all users I dont see a reason to exclude anonymous users. I also agree that for the same person to cast more than one vote under different user names should be considered unethical. (I was not aware that people have multiple accounts, and it is not clear why they do.)

    The second extreme is officers of the blog. Here, to have transparency and accountability, I suggest that we set a definite rule that moderators and administrators should be "real people" that we can clearly identify.

    Voting to close and open is somewhere in the middle. It is not a privilege but rather is an administrative service that lie on the shoulders of users. Here, the votes are not anonymous. People who share the decision to close or open are identified by name and they put their names behind their decisions. Closing/opening/deleting is an administrative call and I think that it should be transparent who made it. I simply propose that anonymous users will refrain from closing/opening questions. **Added later:** In view of the fact that other people dont share my concern, and in light of quid's arguments below I withdraw this proposition about closing/opening.

    If I had my druthers, every MO account would have a full name attached to it, and a link to an academic website.

    I prefer full names too, but the second requirement would knock out non-academic mathematicians, which would be bad.

    • CommentAuthorquid
    • CommentTimeOct 18th 2011 edited

    Thank you to everybody who participated so far in the discussion!

    Mainly for the sake of completeness, as I did not say it so far (though I guess it is implictly clear): Personally, I do/did not see a problem with voting to open/close a question while being not linkable to a 'real person'. The reasons are essentially exactly those already given by several others.

    I do agree that multiple voting (both up/down and close/open) would be a problematic. Yet, specifically for open/close, which we are discussing here, it seems highly unlikely to me somebody would do this; leaving all moral considerations aside, simply the effort required for the effect and the relatively high-risk of being caught (as the votes become visible) to me simply suggests that this a most inefficient attack vector on the integrity of the system. Yet, if somebody should be seriously worried about this, as said, at least I would have no problem, say, to disclose my identity to one or all of the moderators.

    Yet, it seems to me Gil Kalai's concern is of a different nature. His definition seems good to me. However, it is not clear to me which pieces of information that are reasonably obtainable on a somewhat typical MO-user by knowing their identity would be useful in the present context. And, why the act of voting to close/open is so special, as opposed to all other administrative tasks that are fulfilled by users, or also as opposed to discussing about whether a question should be closed/opened (or would this be included). Deletion was mentioned. If I understand correctly, if five accounts with 20 points decide something is offensive or spam, it is instantly gone (except for 10k users), until restored by a moderator (or perhaps 10k+ users). Where is the accountability for the decision here? Of course, if it turns out that these flags were unjustified or even willfully misused the post will be restored (and in the latter case or also in both the accounts be held somehow responsible). Moreover, I would like to add that as far as I understamd (please correct me if I am wrong here) voting up/down is not anonymous; which account voted what is recorded, it is just that it is not displayed to the public. And, this is where I do not understand the argument. Why should there be more need to know which person is behind an account in a case where the action of the account becomes linked to the account visble to everybody as soon as it takes effect. In a case were the closers were not visible I would much better understand the desire to know as good as possible who (real idetity) is in the group of people that can decide this. If I or anybody else should repeatedly vote to close or reopen in strange ways, I think this problem would be much quicker and easier be spotted by the moderators or the community at large than if I would up/down vote in completely strange ways.

    Or, are we mainly talking about courtesy; like, if you close a real person's question you should at least document which one you are yourself. [This is perhaps stangely phrased, I would not want to belittle this concern if it were one, I merely want to understand what the concern is precisley].

    (Finally, since it came-up in passing a response in passing, I can understand and even share some of the arguments in favor of real names. However, for me there are more reasons against them. What can be subsumed as privacy-concerns is not completely irrelevant to me personally, but in the end not at all the main reason. However, in view of my own request, I would not like to elaborate on this in this thread.)


    One more small remark. A newish user whose question is "closed" may feel insulted or slighted. But it is probably worse when votes to close are by "Bugs Bunny" or someone similar.

    • CommentAuthorBugs Bunny
    • CommentTimeSep 17th 2012
    Should we ask people in 200 years who is more "real" Gerald Edgar or Bugs Bunny:-)) ?

    Seriously speaking, this discussion is preposterous. This is an online community and the privileges are based on the reputation earned here. If someone is abusing these privileges, there are moderators to take it away. And there is not yet a single case of voting privilege abuse to report.

    I have already explained my reasons for anonymity in
    and I am not going to divulge any more information.

    Actually, there have been plenty of cases of anonymous accounts abusing the voting system, nearly all of which have been dealt with quietly (e.g. via private email with the moderators). It's never been so much trouble that the moderators have felt the need to restrict anonymous accounts, but I just wanted to be clear that these abuses have certainly happened.

    • CommentAuthorfedja
    • CommentTimeSep 26th 2012
    I do not mind what was called "pseudonymous accounts" at all (some may even argue that I'm using one myself). However, I believe that it is merely a matter of politeness to provide a unique username (i.e., "querty1967" is perfectly fine with me but "unknown(google)" is not). The reason is that I like to see the (virtual) person behind any action that is not merely requesting or supplying some information. It is totally OK with me if the online personality is non-linkable to any real life personality but, IMHO, there should be some "online personality integrity" for everybody using the net as means of communication. I cannot completely rationalize this feeling (no more than the feeling that an open approach with a straight look in the eyes is better than quiet stalking from behind) but that's what it is.
    • CommentAuthorHarry Gindi
    • CommentTimeSep 28th 2012 edited

    On the other hand, I don't think there is any legitimate reason for people to have multiple accounts...