On 2/7/06, David Gerard dgerard@gmail.com wrote:
Consensus doesn't scale.
We've recently seen the numbers from Gmaxwell and Kim Bruning that show that this is not at all the case with articles - except a couple of hundred articles (out of 900k+) which appear to be pathological. *Mostly*, people leave articles to others who know about the area, and those who know about an area mostly manage to thrash out a consensus. The failures of consensus in article editing get a lot of attention but they are the *exception*.
With policy, this hinders change greatly, but it's unlikely to be a major problem in the near future. With wheel warring or serious edit wars, however, the fact that consensus doesn't scale is wasting a lot of our time here. It takes being hauled in front of the arbcom to get any results.
Yeah. It's getting policy consensus to scale that's tricky. Continuous reference to basic principles - "we're here to write an encyclopedia", "NPOV", "don't be a dick", etc - may be a useful touchstone for either deriving a lot of the crufted policy from first principles or getting rid of it.
Those numbers really don't show as much as one might think, though. OF COURSE any individial article will, on average, have only a handful of editors. The problem is that Wikipedia is increasingly becoming large enough to make reinventing the wheel on every individual article untenable. Thus, we get attempts to create policy/guidelines/style guides/whatever centrally and apply them to a (large) group of articles at the same time, which means consensus needs to form not among a few editors of an article, but among all active editors in a field.
This is most obvious in AFD, incidentally; attempts to delete all schools/Pokemon/roads/etc. get a lot more people involved than have edited any single one of the articles in question.
Kirill Lokshin