On WikiPedia, some banned users have repeatedly circumvented their CommunityExile by using new acounts/IPs (cf. LoginsAreEvil). The standard lifecycle for these accounts depends on whether they are high stealth or low stealth:
Low stealth lifecycle:
high stealth lifecycle:
How can we break this ConflictCycle ? A similar point is raised on CommunityExile: that per-user bans make a community psychotic. --MartinHarper
Eliminate the sysops. Instead of concentrating power in the hands of a few, devolve power into the hands of the many. No GodKings. -- SunirShah
If sysops are eliminated, power will rely in the hands of the GodKing and his developers. So is it really sysop elimination, or sysop multiplication which is best ? Or could the masses have power over the sysops by sheer numbers ? --anon.
Use MeatballWiki as an example. Over the past three years, I have worked very hard to DevolvePower from my hands into the community's hands. I'm not done yet. I'm working on a PublicScript. The point is that I'm working on it. Eventually, I want to hand over the reins (or the reign) to a NonProfit corporation. At the very least, Wikipedia can follow our lead as we develop new ways of devolving power, although I suspect that after devolving power becomes a CommunityExpectation the community there will push for new powers on its own. -- SunirShah
I've been formulating a plan for sysop elimination all week. It's almost ready for presention... -- StephenGilbert
Do you have a specific goal in mind? --anon.
Yes, I do. I believe that, without exception, all WikiPedia sysops (I am one of them) have good intentions and want what is best for Wikipedia. However, many sysops wield their power in a way that breaks the trust of those who don't have such powers; AssumeGoodFaith is breaking down on both sides. Part of the problem is that we've become a bit confused; the need to perform certain functions (PageDeletion, for example) is often seen in terms of "we need sysops". If we can figure out how to perform these functions on a community level (e.g. KeptPages), we wouldn't need a class of Wikipedians with super powers. Well, except the developers... -- StephenGilbert (edited)
I was thinking this morning about the life cycle of discussion forums. I observe that over the course of a few years, most of them either fade away (if they lack a critical mass of users and discussion), or deteriorate. The deterioration is usually characterized by a declining signal-to-noise ratio, with the truly knowlegable individuals discontinuing their postings once things get bad enough, often choosing to StartAgain.
The CypherPunks are an excellent case to study because the quality started very high, and at a certain point deteriorated rapidly; that community has been nothing more than a vacuous source of political rants for quite some time. Early articles were well researched, carefully organized, and thought provoking. Perhaps it went wrong because it gained a measure of popularity thus failed to LimitTemptation. Yet, there are other communities that were more successful for longer periods despite having had greater interest to a broader group of people (e.g. GrazersEdge?).
Two common traits among the longer lasting forums are the presence of strong CommunityExpectations and some sort of technological and human capacity to have BuildingJanitors. Wp has both and perhaps this is what has allowed it to withstand the degree of popularity it has achieved without imploding.
More heavily moderated forums usually follow a different path of deterioration in that the best contributors usually leave, one at a time, following a disagreement over moderation practices. It is ironic that even experienced participants, when acting as moderators, have trouble determining which posts will hurt the community and which are necessary to allow it to grow. Such forums die a boring death of attrition rather than developing the content rot that occurs in unmoderated communities.
It is unclear to me where Wikipedia fits into all this. In a way it has been in tranisition constantly since its inception, or at least for the last year or so. I would characterize it, at present, as a fairly heavily moderated forum, so perhaps it will ultimately follow that path.
The best answer to the popularity problem is to LimitGrowth. But this doesn't appeal to many people, so you need other solutions. WikiLifeCycle talks about the problem you identify here, in the context of wikis - but CommunityLifeCycle would be more useful. Not all communities decline, though: some go in cycles of expansion and contraction - especially where the medium acts as an inherent limit on scalability: such as mailing lists.
The stuff at the top needs working into CommunityExile as a section in how people work around their exiles, I think. However, I don't believe the failure of exiling on wikipedia is due to the need to DevolvePower, as Sunir suggests, but rather due to the need to have FairProcess, as Steve suggests. HTwoGTwo has a handful of GodKings who can ban without appeal, but they also (belatedly) have FairProcess, so they get away with it. The same applies to ISP bans of those who breach their TOCs - their is a centralised decision-making authority, but it follows FairProcess.