14

Excerpt from https://en.wikipedia.org/wiki/Wikipedia:Why_Wikipedia_is_not_so_great

Editors have learned that formation into "gangs" is the most effective way of imposing their views on opposite-minded contributors. It makes a travesty of the revert-rule when one individual can simply send an e-mail alert to friends requesting a timely "revert favour" once he has reached the limit of his daily reverts. This may apply to deletion debates as well, where a group of editors may be organised so as to always vote en masse in favour of keeping an article written by one of the gang, or related to the gang's main field of interest; or to push through deletion if their interest is a deletionism. Gangs sometimes do serious damage to Wikipedia's policies and guidelines also; by ganging up they can be written to say almost anything.

Summarized, Wikipedia suffers from cabal-like behaviors, in which controlling narrative is a relatively straightforward process. In short:

  1. Choose a topic to control.
  2. Find or hire like-minded users, and coordinate with them via a private network for immediate coordination.
  3. As objections to edits typically filter in one by one, coordinating with the rest of the cabal, one can claim Wp:Consensus even if the aggregate is actually opposed to this.
  4. Appoint cabal users as administrators, to assist with "problem users". Abusing this privilege is a relatively safe process as the essay also states:

Wikipedia administrator vandalism itself is only controlled weakly, and there's insufficient power to desysop a popular tyrant. Only the most abusive administrators – perhaps 2% total – have their statuses removed.


Is it the case that SE has or actively does suffer from similar issues? Say for example, would it profit Foo Incorporated, a competitor to The Bar Foundation, to hire a few people to actively monitor their respected tags, looking to harass Anti-Foo contributors, questions, answers, and tags while promoting all things Pro-Bar?

On the various SE's, I have never personally recognized any behavior that would pass for shilling, but maybe I am not looking hard enough.

If it largely has not been an issue, to what can we attribute this success to? If it has, where and why does it exist?

  • 2
    No clue if it happens or not, but if it does, it likely gets picked up by moderators pretty quickly. R/A comments that get flagged or abusive voting patterns usually get attention – Olivia Dec 14 '18 at 6:44
  • 4
    @PrincessLuna Not necessarily. There was a very offensive and dismissive post on another SE site which stayed up more than 6 months. My first flag was declined (as were several other peoples') because the offending user was popular. Only just recently did it get removed after I flagged it again. The issue is not only with moderators, but with users who have a high reputation. Note that this is not a network-wide issue at all. – forest Dec 14 '18 at 10:43
  • Note that the page you linked to is an essay, not a policy, nor a guideline. It is an opinion piece written by some editors. I would tend to disagree with your statement that "controlling narrative is a relatively straightforward process" on Wikipedia, especially the bit about the abuse of administrators. – isanae Dec 14 '18 at 10:43
  • 1
    @isanae I actually agree with it fully. There are lots of issues with abuse of power on Wikipedia that occur to this day, especially on controversial topics. – forest Dec 14 '18 at 10:44
  • @Akiva How would you know? You're the one who came up here with this essay, the burden is on you, not on me. You're quoting this essay as if it was the truth, or at least widely accepted. The fact that it stayed an essay instead of becoming a guideline suggests that there is no consensus about the subject. Most of it, especially the "Miscellaneous" section, is just a pile of crap. – isanae Dec 14 '18 at 10:58
  • 1
    @forest And that is not an understatement. The nuances surrounding abuse are so brilliantly executed, that you have to be fairly sophisticated to fully comprehend what happens to people targetted there. Defend yourself, and you are guilty of not taking responsibility for your actions. Take responsibility, and you will be accused of trolling the administrators because you were not being genuine. Suggest the administrators are biased, and you are guilty of not assuming good faith. Claiming the admins are not assuming good faith on your behalf, and you are guilty of ... well you get point. – Akiva Dec 14 '18 at 11:00
  • @Akiva Well I think it's more subtle than that. I don't believe any kind of conspiracies or malice is involved (usually, at least). It's just natural human behavior to e.g. circle the wagons, defend your own circle or beliefs when you have the power to do so, etc. – forest Dec 14 '18 at 11:02
  • @forest It all depends on consequences of the opinions at bare. The content regarding the History of Bergundy would really only put the ego of historians editting the article on the line. The content regarding Russian_interference_in_the_2016_United_States_elections has potentially the entire democratic party's future on the line, especially when you consider the most plausible alternative explanation stems from Assange inferring that their leaker was the murdered Seth Rich. If you do not think people are consciously invested in making sure that discussion is not had, I'd say look closer – Akiva Dec 14 '18 at 11:22
  • 1
    @forest I said likely. There ARE a ton of posts that slip through moderation, that includes all types (questions, answers, comments, chat messages). Most of it gets caught, there will still be exceptions (caused by aging away, a bad review, etc). Likely != all – Olivia Dec 14 '18 at 11:24
  • @PrincessLuna It's more common than not in certain cliques. – forest Dec 14 '18 at 11:31
  • 2
    @Akiva Oh for the highly political articles, I absolutely agree. I recall a study showing that Wikipedia had a stellar reputation for accuracy for the hard sciences, but a very bad track record for controversial politics. – forest Dec 14 '18 at 11:32
  • 1
    @forest if you have proof of moderator abuse, send stack exchange an email. – Olivia Dec 14 '18 at 11:33
  • 2
    @PrincessLuna I have. Several others (including moderators) have as well. Not that I'm here to complain about that. I was just making a comment for my own anecdotes. – forest Dec 14 '18 at 11:35
  • 3
    It's too late. Russian bots own SE so hard that they got their own version of SO. – Won't Dec 14 '18 at 16:14
14

I'd say in a "soft" way, like arguing strongly together for a field of interest like a tag somewhere on a site, yes. That could certainly happen here.

But in the more problematic ways you could imagine, like coordinating R/A / Spam flags to nuke undesirable posts quickly and without consensus, or sockpuppet voting to inflate member's scores, those get picked up usually by the moderators. With respect to the process once that happened, it varies I think, but generally the mods will come down like a ton of bricks on anyone trying this kind of stunt.

SE has automated as well as manual tooling to detect voting anomalies, so coordinating up and downvotes over time tends to not go well and end in account destruction or suspension of those guilty of it.

Especially this:

Find or hire like-minded users, and coordinate with them via a private network for immediate coordination.

I guess could work if you limited it to comments and close votes, perhaps errant downvotes. But once you start actually coordinating for voting and flagging, it's highly likely that it'll draw the suspicion and ire of someone. And just one person who sees the pattern is enough to kick off a full-scale moderator investigation via custom flag.

Appoint cabal users as administrators, to assist with "problem users". Abusing this privilege is a relatively safe process as the essay also states:

This is directly made impossible due to the fact that SE holds moderator elections, or moderators are appointed via CMs. The CMs have their own selection process and are employed by stack exchange, so you can't really influence that too much. Moderator elections are site-wide, so unless you can muster enough support to actually sway majority opinion, that won't happen either.

Wikipedia administrator vandalism itself is only controlled weakly, and there's insufficient power to desysop a popular tyrant. Only the most abusive administrators – perhaps 2% total – have their statuses removed.

Another strength of the SE system of moderation comes into play here. Moderators are highly accountable to each other and the CM team. Trying to pull a fast one like that as mod here would realistically see you lose your diamond near instantly.

On top of all that, there are users who look for this kind of stuff just for the fun of foiling it, so operators of such cabals are always opposed.

  • 3
    Not to mention, mods here don't have the same job as admins on Wikipedia. Mods here tend to deal with abuse, not "fact checking" or ensuring individual posts are accurate (which is often highly subjective). – forest Dec 14 '18 at 10:45
  • 1
    @forest Maybe so... but even then, what is considered "abuse" itself is subjective and may be used as a way of silencing opposing opinions. – Kevin B Jan 30 at 21:10
4

I think it could happen, but as a moderator on two sites (one medium-sized, one small) I have only ever seen weak evidence for such cabals starting to form and they do not seem to last long. I think they only extend as far as the first two points in your list.

Neither site has a strong chat culture, and this may mean that most users act independently most of the time.

I think a strength of Stack Exchange sites that helps reduce the tendency for cabals to form is that most of the sites' users focus on on-site content rather than off-site contact.

If a fully fledged cabal were to emerge, and I detected it, I have confidence that the powers afforded to moderators, many of which I have never needed to engage, would be able to contain, and then disperse it.

  • 4
    users focus on on-site content rather than off-site contact. what does this mean exactly? How is this different than say wikipedia? – Akiva Dec 14 '18 at 7:17
  • @Akiva I think it would be very similar to Wikipedia in that respect, but if either promoted social network activities then both would suffer more than they do. – PolyGeo Dec 14 '18 at 12:04
4

Its somewhat less likely to happen on SE simply because of the structure of things.

Sufficient numbers of users with enough reputation can undo most of what I can do.

As for your specific stages...

Choose a topic to control.

There's enough subject matter experts to call someone out on incorrect information that its mitigated. I get schooled a lot

Find or hire like-minded users, and coordinate with them via a private network for immediate coordination.

Thats... slightly tin foil hattery. That said, its been known to happen accidentally and some sites have had massive voting corrections. There's something to be said for stack exchange's chat being default public, and most mods tend to keep mod business on the site.

As objections to edits typically filter in one by one, coordinating with the rest of the cabal, one can claim Wp:Consensus even if the aggregate is actually opposed to this.

I can't really think of an equivilent here. In theory if something fishy is happening involving mods and edits and not really much on meta, I guess someone could let the CMs know

Appoint cabal users as administrators, to assist with "problem users". Abusing this privilege is a relatively safe process as the essay also states

I can't really appoint anyone as anything other a chat RO. On SE, you can either get elected by the community as a mod, get appointed as mod on a younger site, or get reputation. Its a pretty solid meritocracy. While I guess a popular mod has a lot of influence we also are open to getting questioned. I'm sure a series of actions pushing an agenda would be questioned.

2

Yes, this happens on Stack Exchange. While we like to think the voting system works perfectly as intended, it is only very well executed.

There are several Stack Exchange sites that work on a subjective basis, because the questions are necessarily "soft" and the answers generally based on subjective personal experience rather than objective data.

The result is some communities run by cliques that have by-and-large determined a particular type of answer that is acceptable, and a type of answer that is not; more there is a type of user that is considered worthy and others that regardless of reason, are treated with disdain.

Even on the sites with objective Q&A, cliques designed explicitly to downvote and delete main posts viewed as unworthy, and target meta posts that dispute this approach, especially where this is far from being a consensus. Math SE's CRUDE is considered an example of this by some.

This is not surprising at all. While Stack Exchange has higher standards on paper, they are still being applied by people. Better people than the average internet user, but people who have special interests and a tendency to Other just like any.

You must log in to answer this question.

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