Yesterday I spotted that on stats.stackexchange.com questions that request analytical solutions of statistical problems and answers that give nicely derived answers using mathematical calculus are not distinguished. I think it's useful to see a collection of mathematically derived answers while many other answers are opinions, collections, applications of well known methods, numerical approximations etc. I created the analytical derivation tag, wrote a tag description, went through about 400 questions and answers, selected 60 that qualified, and tagged those that had not had already 5 tags. After receiving the taxonomist badge I went to bed.

Today I wanted to check if users received tag-badges who were particularly active asking such questions or giving such solutions and I wanted to continue tagging, however I had to see that all of my several hours of work has been reverted! This has been done without any discussion, not even a bit of explanation!

I view this as ruining the results of several hours of salutary efforts and given that the nullification of my work was done in secret (not contacting me, not discussing, not giving explanation, not suggesting a way to improve (if the destruction was related to a presumed imperfectness at all), not even raised on the stats meta) – I believe this was an abuse with their rights of a moderator.

Please tell me how I can get my work restored, how I can continue with it, and how to go on with the moderator overriding their commission.

  • 25
    $\begingroup$ Such big retaggings and changes should always be discussed on the appropriate meta before you apply them. $\endgroup$ – Bobby Apr 29 '11 at 11:15
  • $\begingroup$ @Bobby Would you please underpin your opinion? $\endgroup$ – GaBorgulya Apr 29 '11 at 11:20
  • 5
    $\begingroup$ I agree with @Bobby. Stuff like this needs to be discussed. See e.g. the Meta retag-request tag: meta.stackexchange.com/questions/tagged/… $\endgroup$ – Pekka's other trolling account Apr 29 '11 at 11:25
  • 6
    $\begingroup$ Wouldn't this discussion be more appropriate on the stats meta? I assume the moderators responsible could give you their opinion, as well as the users whom it directly affects. $\endgroup$ – Cody Gray Apr 29 '11 at 11:26
  • 4
    $\begingroup$ @GaBorgulya - When you edit many questions like this you affect their last activity date, sending them to the home/active pages for all to see, it's very disruptive. A moderator can do this much more efficiently without that disruption, and as others have said it should be discussed on meta. The discussion gives the community a chance to discuss editing tons of questions, rather than one person making that decision. $\endgroup$ – Nick Craver Apr 29 '11 at 11:32
  • $\begingroup$ A stats moderator may continue restricting the discussion on stats meta. The question of abuse can not be raised on a forum where they have distinguished power. $\endgroup$ – GaBorgulya Apr 29 '11 at 11:34
  • 3
    $\begingroup$ @GaBorgulya - Moderators cannot delete content where we (the dev team and other moderators) cannot see it. If a moderators is abusing the system trying to do that, then report it here and I assure you it'll be dealt with. $\endgroup$ – Nick Craver Apr 29 '11 at 11:36
  • 6
    $\begingroup$ That's a pretty serious accusation. There's a big difference between restricting discussion and rolling back mass tag edits. If or when that does happen, then you should come here and report it. But it seems like you should give the mods responsible a chance to explain themselves, before accusing them of malfeasance. $\endgroup$ – Cody Gray Apr 29 '11 at 11:36
  • 5
    $\begingroup$ Can you prove it you are not abusing the taxonomist badge? $\endgroup$ – YOU Apr 29 '11 at 11:37
  • 1
    $\begingroup$ @YOU Innocence does not have to be proved. But if you question the usefulness of my efforts, please reflect on my first paragraph which gave some ideas why I think it was useful what I did. $\endgroup$ – GaBorgulya Apr 29 '11 at 11:41
  • 3
    $\begingroup$ I've moved this to meta.stats; this should first be discussed in the context of the site $\endgroup$ – Marc Gravell Apr 29 '11 at 11:43
  • 2
    $\begingroup$ @GaBorgulya, you said "you went to bed after you got that badge". Looks like that was your goal. $\endgroup$ – YOU Apr 29 '11 at 11:45
  • 2
    $\begingroup$ @S.Mark My motivations were: 1. doing something useful for mankind (even in my unpaid free time), 2. yes, earning the badge (badges come after significant efforts, also their intended purpose is to give some additional motivation), 3. I'm admiring the nice analytical answers of users like @probabilityislogic, @cardinal, occasionally @whuber, my work is also towards the recognition of their efforts with corresponding tag-badges. I don't see anything negative in my motivations. Earning a badge is a good signal to go to bed at 2:30am after enthusiastic work. $\endgroup$ – GaBorgulya Apr 29 '11 at 12:03
  • 5
    $\begingroup$ I can understand your frustrations, but I feel the need to object a little to your framing of this question. Certainly this is a valuable discussion to have on Meta, but I wish it hadn't started with an indictment of the moderators. Don't forget that they've earned their positions through many hours of their own "salutary efforts". Even if you think they were wrong I think they deserve some respect, consideration and the room to make honest mistakes. I am willing to believe that you thought you were acting in the best interests of the community. I am willing to believe that the mods were too.. $\endgroup$ – JMS Apr 29 '11 at 20:17
  • 3
    $\begingroup$ Incidentally, I found the wholesale retagging disruptive (lifting all kinds of old questions to the front page) and support remedial action of some kind. $\endgroup$ – JMS Apr 29 '11 at 20:23

I am truly sorry that all that work has disappeared and understand the shock you must have felt when that happened.

Such a result actually is a foreseeable consequence of any large-scale,unilateral action taken on an SE site. These sites are community run. That does not mean that individuals do as they will and then ask that others accept the consequences. It means that members work together, collaborate, talk, test things out, and move forward with consensus. Meta and chat are two of the (many) mechanisms to support these activities.

I instigated the response to your actions after (a) rejecting two requests from you for a wiki edit on your new "analytical-derivation" tag (the description was vague and incomplete) and (b) receiving a strong complaint from a highly respected member of the community about the wholesale retagging. Those actions included:

  • Writing to the other stats mods about the retag-fest. (It was very early a.m. their time so no immediate response was expected.)

  • Consulting with mods on other sites and SE representatives via chat.

  • In response to their (unequivocal) advice, contacting the SE team by e-mail.

In the email to the team I gave the facts: we had an active, valued user who created a new tag and systematically retagged questions until he was notified of the "taxonomist" badge, then stopped. I then shared my own impression:

Sounds like abuse to me, although fairly harmless to the site. The tag would be potentially useful but IMHO has now been applied so indiscriminately that it's pretty worthless now.

Ideally I would like to have most of this tagging reversed but that could alienate this guy, who's pretty active. I'm also wondering how to head off such abuse in the future.

Finally, I did not specifically request any action:

I look forward to any suggestions you might have to offer.

I did not want to make a unilateral decision because that would be unfair to you and the community. I did not write to you because I felt that was premature: I was still seeking advice and guidance.

The moderators cannot (easily) remove a tag: I believe that was done by the SE team. It is an outcome I thought was likely when I wrote them, so I take responsibility for it. The message you received (from another mod) was a courtesy to let you know what had happened, why, and what to do in the future. If now, knowing more of the history, you have constructive suggestions for how I could deal with similar situations in the future, please let me know: I needed help with this one and welcome help from anyone at any time.

I think I can speak for many in saying we value your interest in and engagement with this site. That's not just empty words: in my opinion, the quality of your answers is increasing and they have become regular and welcome contributions. However, as a relative newcomer, your experience with this community is (necessarily) limited. I'm sure you felt that what you were doing was another contribution, but due in part to this ignorance of how a community works, your efforts backfired for many reasons made abundantly clear on the current thread. Unfortunately, your subsequent actions do you little credit: in the process of seeking to understand and correct what happened, you (a) indiscriminately attacked the moderators of this site and (b) did so outside the usual forums provided for such discussion.

Your "several hours of salutary efforts" have not been wasted if it has helped you understand how this community works and how you can make constructive contributions in the future.

  • $\begingroup$ I will definitely draw my conclusions from this issue. I also suggest you moderators doing the same, and I highlight that when deletion of work of many hours suddenly occurs without any explanation in an impersonal way that is very frustrating. I still see my yesterday efforts useful and the "damage" label on them wrong-headed. No doubt, the label definition can be improved, a couple of labels cast may need revision, and maybe a better expression could be found for a similar topic. But there is forum for discussion, there are methods to rename tags. The constructiveness is what I lack, too. $\endgroup$ – GaBorgulya Apr 29 '11 at 14:28
  • 6
    $\begingroup$ @GaBorgulya, I am sorry that your work was lost, but this happens more often than not, when somebody starts doing something that nobody else asked whether the end result is useful or not. Note that your accusation that many hours work was lost without notification goes both ways. A lot of hours were spent to get the site as it is now, introducing major change unannounced and neglecting that work is also frustrating. $\endgroup$ – mpiktas May 3 '11 at 8:20
  • $\begingroup$ @mpiktas The case is far further than this now. There was a second thread to restart the tag from scratch with the required discussion. Mbq there declared s/he viewed me as malicious. After 3 days s/he did not rethink that. I'm now waiting for the answer of the SE team and that will decide if this will be my last post to stats.SE or not. $\endgroup$ – GaBorgulya May 3 '11 at 8:30
  • 11
    $\begingroup$ @Ga You are starting to move beyond the pale. A search of "malicious" in the two meta threads pertaining to this question shows three uses, all by you, not by @mbq. Making scurrilous (and false) accusations and repeatedly making threats are unconstructive and antagonistic. As a moderator, I will feel obliged to remove any more such comments, whether made by you or anyone else, without any more warning or explanation. $\endgroup$ – whuber May 3 '11 at 13:54
  • $\begingroup$ I wish you success in running stats.SE in the future. Regards $\endgroup$ – GaBorgulya May 4 '11 at 8:28

While I sympathize with your efforts and have no doubt about your good intentions, I would like to draw your attention on two points.

First, it would have been better to propose such big changes for discussion here, on Meta. This is the ideal place for discussing whether adding a new tag would be relevant or not, in the spirit of this question Regression tags, for example. It appears that such a tag might lead to some confusion, and while you coined the term "mathematically derived answers" to justify it, it's not obvious (to me at least) how it can be used in a search engine.

Second, I would have appreciated that you posted this question on our Meta, rather than the SO Meta because we, users and mods, are the first concerned.

I often found myself spending 2 to 4 hours (day and/or night) on stats.SE because I simply like the idea of sharing thoughts and experience on a wiki. Whenever I can contribute to it, I try my best. But be sure I'm not after rep, badges or whatever. I simply find it is a better opportunity to learn, exchange, and archive a lasting record of great solutions to specific questions, compared to mailing-lists. That's the purpose of the SE community, unless I missed something. So, this is what I was doing as a "simple" user, and that I try to keep going as a mod. Being a mod also means that I have to pay attention to how the site is going on, but I always try to think as a user, and when I'm flagging questions, answers, or when I make some edits, it is has an user, not a mod. Before closing a question or an inappropriate reply, I generally leave a message and wait for user action or others' opinions. We are not "super-user", we are not here to control everything, because this site is driven by its community of users. Frankly, I don't think there was any abuse from our moderation privileges in this particular case. As I said, it would have been better if you asked us before, and when I say us it means all users.

Also we are fortunate enough to have protected a respectful, though punctilious, place since the beta stage. I hope this will continue, for the good of all.

  • $\begingroup$ I was granted the "create tags" privilege at the reputation of 300 and the instruction was "How do I create a new tag? Simply enter a new tag with your question and it will be created". You emphasise that I should have discussed before introducing a new tag, however that was not a requirement. On the other hand I believe you should have discussed before deleting. It seems that you deleted because I did not discuss, and I wrote on the meta because you did not discuss. $\endgroup$ – GaBorgulya Apr 29 '11 at 13:17
  • 3
    $\begingroup$ @GaBorgulya It is a very unfortunate situation because it seems those two messages crossed at some point. Anyway, there's a difference between creating a new tag for your question, and creating a new tag for retagging older ones. That's the point that deserves some meta discussion, IMO. $\endgroup$ – chl Apr 29 '11 at 13:29
  • $\begingroup$ I consider myself unselfish (most people here?), maybe I'm a bit less unselfish than you are. I also answer questions because I enjoy doing so and to help the askers. However, just like scientific publications I'm ready to show my SE contributions if it gives me advantage - I have worked for them. I also find badges useful, a good example is "electorate" it's the easiest gold one, but it motivates to do something useful: to review hundreds of past questions and answers. I'm not ashamed of having the motivation to earn it, I learn andthe community also benefits from users who reviewed the past. $\endgroup$ – GaBorgulya Apr 29 '11 at 14:58

Here is a copy the private message from the moderator that was sent after my post to the meta.


I'm writing in reference to your Statistical Analysis - Stack Exchange account:


It seems you have added a tag "analytical-derivation" to many questions. As you can see, this tag has been wiped out. There are two reasons why we did that:

For the future, if you think that some global editing action is required, propose it earlier as a meta thread.

Regards, XXXX Statistical Analysis - Stack Exchange moderator

I'm not sure the moderator wants to identify themselves so I substituted the name with XXXX.

  • $\begingroup$ Re: meta-tag I'm not sure this is a meta-tag. The numerics tag also exists. I would call that numerical approximation instead and that is a complementary tag to analytical. I think of both as valid and non meta. Even if somebody thinks that it is a meta-tag that should be subject of discussion, not immediate deletion. $\endgroup$ – GaBorgulya Apr 29 '11 at 12:30
  • $\begingroup$ Re: "only way" There are several examples of systematic changes of tags on SE, especially replacing and merging tags. There are several examples of wanted and needed conscious "taxonomic" decisions complementing "organic growth". $\endgroup$ – GaBorgulya Apr 29 '11 at 12:41
  • 1
    $\begingroup$ No need to XXX when quoting them. The only reason I was selective is that I couldn't quote anything etc, and they weren't around (we sleep, occasionally; not often) to add context. $\endgroup$ – Marc Gravell Apr 29 '11 at 13:17
  • 8
    $\begingroup$ It was me and I have nothing to hide; and I was writing this message without knowledge of your meta.SO question, since I have better things to do than spying you across the network. $\endgroup$ – user88 Apr 29 '11 at 13:23
  • 4
    $\begingroup$ Meta tag describes question, not the question topic. [numerics] group questions about numerical methods usage in stats, ml, etc. [analytical-derivation] group questions that require analytical answer. $\endgroup$ – user88 Apr 29 '11 at 13:23
  • $\begingroup$ @mbq - oh, I was wrong then; I thought I saw a record that this was done by a developer. If it was you, then no problem. I'll go away ;p $\endgroup$ – Marc Gravell Apr 29 '11 at 13:24
  • $\begingroup$ Finally, those "taxonomy decisions" are just a janitorial work, and if not, there are discussed on meta. $\endgroup$ – user88 Apr 29 '11 at 13:25
  • $\begingroup$ @Marc No problem; thanks for handling this on Meta. $\endgroup$ – user88 Apr 29 '11 at 13:26
  • $\begingroup$ You answered only after my post to the meta, deleting was before my post. No, you should not be "spying [me] across the network". You should discuss before deleting the work of somebody even if you believe it's of questionable utility. And even if you don't discus, an absolute minimum is that you give an explanation when you delete something. The expected SE behaviour is to give explanation even for a down-vote – deleting many hours of work has a much larger weight than down-voting. $\endgroup$ – GaBorgulya Apr 29 '11 at 13:33
  • 4
    $\begingroup$ This was imposing SE policy, thus the rush was to minimize the damage. I agree that the message to you come too late, however I assure you that it was sent ASAP, as the tag deletion requires action of a SE developer and we live in which timezone we live. And I repeat that I have no knowledge of your Meta.SO post, since I checked only Meta.CV before sending it. $\endgroup$ – user88 Apr 29 '11 at 13:48
  • $\begingroup$ There were questions that explicitly asked for deriving formulas. Some questions received both analytical and numerical solutions. Reading through multiple numerical solutions may give the key idea to solving a new problem. The same applies to calculus used in statistical problems. $\endgroup$ – GaBorgulya Apr 29 '11 at 13:48
  • $\begingroup$ @mbq Although you say you sent your message ASAP you left me sufficient time to worry what had happened, to check my posts and the tags, and to write a long post to the meta, re-word it and submit it. I accept that you did not know about my meta post before your message. I however reject labelling my work as "damage" that had to be "minimized". $\endgroup$ – GaBorgulya Apr 29 '11 at 14:03
  • 1
    $\begingroup$ @mbq I find the tags feature of SE very useful, and using the right tag often needs substantial statistical knowledge. I find it sad that a moderator can think about arranging tags as "just a janitorial work". $\endgroup$ – GaBorgulya Apr 29 '11 at 14:42

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