MeatballWiki | RecentChanges | Random Page | Indices | Categories

When people realized that others were using their moderation privileges on SlashDot for less than altruistic purposes, some means to introduce PeerReview was requested. Since moderation is only granted to those with some karma, removing the abusers' karma is one quick way to admonish and diminish the abuse. However, since RandomPeerPrivilege was introduced to get around the abuse and unscalability of the hand-picked PoliceForce, leaving the admins to suck karma off of abuses was impractical. Consequently, MetaModeration was introduced.

Each day, any user who so decides, can load http://slashdot.org/metamod.pl and moderate ten comments that were previously moderated. They are presented with one of the moderations on that comment which they are to judge whether it is valid or not. If the user finds that the moderation was faulty, say moderating a post as a Troll when it was Insightful, or moderating a troll as Insightful, the user may correct the moderation. A correction takes one point off of the original moderator's karma.

However, after this was put into place, the natural next phenomenon took place. MetaModeration abuse.

Since MetaMetaModeration can recurse to impractical infinities, clearly MetaModeration is inadequate.

KuroShin's answer was to publish the moderators' names in an AuditTrail. This certainly revealed who the stalkers were. (See WhatIsaStalker.)

I am not convinced that the possibility of recursion makes MetaModeration inadequate. Perhaps abuse gets less each step up the recursive ladder. For instance, if there is less abuse on slashdot after MetaModeration than before, MetaModeration has some sort of utility, right? (I have not been watching long enough to compare the pre-MetaModeration situation until now; what do you think?) -- BayleShanks

I believe metamoderations are themselves metamoderated. It's tail recursive, after all, so it can be collapsed into an infinite loop. Still not very good, but not as bad. -- SunirShah

Metamods are not themselves metamoderated. However, the entire population is trusted to metamoderate instead of a small group of moderators, and it requires a certain threshold of metamoderations to apply the punishment for unfair moderation, so a few unfair mods simply don't count. It's a reasonable system, though half-empty simply opened up a simple up/down moderation to every logged in user, with no +5/-2 cap, and it works just fine. You could moderate submissions as well, and choose to read the top-rated submissions. Slashdot's sole mechanism of direct community input strikes me as gimmicky, more than a little pretentious in its "cleverness", and still little more than a thin veneer over the total editorial control exercised by its less than professional editors. They get paid to do this, they pretend to be the mecca of nerddom, so people are inclined to cut them a whole lot less slack ... I think I've veered off the topic though ;) --ChuckAdams

I've written about this in a recent blog posting [Dunbar, Altruistic Punishment, and Meta-Moderation] -- ChristopherAllen

From the article: Finally, if they added to the simulation the ability to punish those who did not participate in punishing (i.e. didn't pay the cost to punish defectors), then the percentage of cooperation that evolved was never less then 60%, and in fact got better as groups got larger.
I'm curious whether there was any cost for this meta-punishment? It might provide a nice theoretical model, but in real world models, the application of punishment does involve cost

CategoryRatingSystem CategorySlashdot


MeatballWiki | RecentChanges | Random Page | Indices | Categories
Edit text of this page | View other revisions