Wikipedia talk:GNE Project Files/Proposed GNU Moderation System

The whole area of reputation is problematic - hierarchy is the result of poorly thought out delegation systems. wikipedia does fine with informal negotiations via the "talk" meta encyclopedia. A formalization of this, via a consensus process (where no one person but perhaps any two people registering a reason or objection could remove an entry), seems more useful than an elaborate tangle of signatures, permissions, promotions. "Any two" could be expanded to a whole system of filters, where the reader not the editor chooses whose diffs to trust. Strong moderation style can be enforced by ruthless appeals based on the "reason or objection", and having articles spend some time "in the doghouse" before they are (automatically) promoted to visible... all of which is lower overhead than the system as proposed above.

Some systems like this were discussed by the United Nations University years ago - I believe the 1993 State of the Future report of the Millenium Project (of the American Committe for the UNU http://acunu.org) addresses this issue in some depth, based on the assumption of variant points of view. Not sure what was done about it.