The short reason why I stopped editing at Wikipedia is simply that it was no longer any fun. Scanning my watch list, reverting vandals, cruft patrol, and troll abatement had become just another duty that had to be performed in a duty-filled day. Creating new content, the reason I joined the project, had become more the exception than the rule. Now that I have been away from the wiki for a few months I've had time to gather my thoughts on just why I had to leave and why I won't be going back anytime soon.
As Jason Scott put it:
"...the content generators stop being so and have to become content defenders."
the reason for this is that the more times one contributes a new article or makes significant contributions to existing ones, you become a stakeholder in those articles and by extension to the community of editors in those categories, and to the Wikipedia project as a whole. Thus attacks on the integrity of an article, become attacks on your integrity as an editor and the editing circles that you belong to. Defending content from poor or malicious editing is in fact a defense of the relevancy of your own contributions.
Note I used the term 'stakeholder' rather that 'owner' and these shouldn't be confused. A stakeholder enjoys some of the benefits without the privileges of ownership but only by assuming most of the duties. The frustration arises when you find that despite this sense of responsibility, and despite the effort you have made, Wikipedia does not value you above some random anonymous first time contributor, or indeed any editor already under sanctions for bad behavior in the recent past. The stock answer is that Wikipedia has been constituted as egalitarian effort and rules and policies are in place to deal with malicious and unacceptable editing; but therein lies the problem, the rules and policies are very hard to enforce. In fact they are exhausting to enforce and that is why conscientious editors leave. That is why I left. Worse, the more effort that is made to defend the integrity of the Project, the more valuable a target for exploitation by malicious editors it becomes.
Surprisingly (some would say perversely) Wikipedia is seen as reliable because many of the traditional authoritative channels are perceived to have been compromised or not updated frequently enough in areas subject to rapid change. For example, while it is premature to declare that the peer-review system in traditional journal publishing is in trouble, the fact remains that there have been some spectacular and very public failures. If we add to this the fact that the formal language and specialized vocabulary of academic papers may be obscure to those outside the field (in fact in some cases obscure even to those only a few degrees removed from the subject) and take into account as well the appearance of several journals with all the trappings of a traditional publication, but covering areas well outside the mainstream, it is unsurprising that mere publication in and of itself is no longer seen as a guarantor of legitimacy. While arguably those in the fields in question can make a distinction when they need to, there exists a constituency of those whose opinions carry weight that may not. Also, the Internet has made it possible to appeal directly to the general public for funds in a way impossible in the not so distant past, ether by the sale of books or by donation. In both cases these people have been turning to Wikipedia, first because by its nature an encyclopedia distills complex subjects down to a explanations that are accessible to a wide audience, secondly because Wikipedia's system of self-policing is widely viewed as less subject to being biased than those of other sources. This being the case, any idea that is seen to survive the process immediately gains status and a cachet of legitimacy that it might not otherwise have if it was limited to some obscure journal, or was being disseminated from a single website. As a bonus Wikipedia's many mirrors raises the Google rank of key terms, also adding to the ideas perceived importance.
This of course is not limited to mountebanks, there exists a class of editor so driven by ideological agendas that they simply will not recognize Wikipedia's Neutral Point of View Policy or seem to believe that it means that it guarantees uncritical place for their interpretations regardless of how flimsy the supporting facts or underlying logic might be. Worse, after an exhausting effort to bring these under control in a few months a fresh batch of POV pushers, unrelated to the first, show up to the same topics and the process must begin again from scratch. There is also the problem of the more traditional (read: deluded) crank with pet theories and a very superficial understanding of the field, to say nothing of their disciples. A few months before I left I was treated to the spectacle of no less than six editors claiming Phds trying to reason one of these idiots out of his notions of the existence of a ceramic gas, and thinking what waste of talent. In the end the crank had to be brought before ArbCom and was subsequently barred, but only after tying up mine and several other editors time for months. Undaunted, this individual has opened several sockpuppet accounts and continues to push his ideas on the same pages he was barred from. The fiction is that these people need to be educated in the ways of the 'pedia - the truth is that by in large they are beyond redemption because they are parasites, scofflaws or insane.
Passing mention of other irritants include those that feel the need to 'polish' otherwise stable articles with bad grammar and oversimplifications; editors for whom English is a second tongue but have no grasp of this language's idiom making a stand on what they perceive the meaning to be; link spammers, viral marketeers, and astroturfers; and of course the constant drizzle of schoolboy vandalism.
Now this will probably be seen as the usual parting rant of a burnout with an overinflated idea of his own worth and the importance of his contributions, and I'm willing to admit there might be some truth to that view. The grim fact however is that the sandbaggers (because that is in essence what we are) are losing ground because the burnouts are not being replaced in the ranks of the content defenders fast enough to offset the influx of bad editors. Over my last last six months as an active editor it has only been by each of us taking on more work that we have kept our heads above water in the science and technology sectors alone. It is glaringly apparent to me that eventually these matters will be addressed simply because there will be no other option, the only questions are how much damage will be done in the interim and how draconian will the new policies have to be.
I have read some of the suggestions for change that have been put forward, by in large these tend to throw the baby out with the bathwater. Wikipedia is what it is because of the rules and policies now in place; that means all the good along with the bad, and despite the issues discussed here there is far much more good. Thus ideally, any changes should have a minimum impact on regular users and editors, yet allow the worst of the problems to be brought swiftly under control. One thing that works very well, albeit slowly, in dealing with problematic editors is ArbCom. Naturally as a court of last resort that is involved largely with serious charges of rule-breaking, cases that wind up there are complex and thus require careful and lengthy examination of the evidence. However all of the conflicts that I have been involved in, that wound up at ArbCom, started as content disputes that escalated. Looking back on many other cases that have gone to ArbCom it's apparent that this is the situation in for most of them, and in the overwhelming bulk of those there was an apparent violation of basic policy, like one of the Three Pillars, or What Wikipedia is Not. Had evidence been presented then and there a ruling could have been made and it would have been over. Some of these were clear issues of NPOV violations, yet the bickering went on for months until it got to the point where behavior problems broke out and it was on these that it went to arbitration.
I submit that the present system of dispute resolution is is quite simply overwhelmed which results in disputes escalating far beyond the point where they need to be, thus becoming far more complex to sort out than they have to be. The solution, as I see it, is to create a much larger pool of arbitrators who would accept cases earlier in the conflict, and expand the purview of arbitration to include violations of basic policy.
The big advantage to this idea is that it leaves the general structure of Wikipedia in place. Other advantages are: It can be implemented on a trial bases, if it doesn't work out the whole thing can be dissolved at the whim of Jimbo Wales and things will go on as before; it will reduce the impact of disruptive and malicious editors before they cause an escalation of conflict (almost necessary now to get them in front of ArbCom); it will support and encourage the sandbaggers that are trying to defend the integrity of the wiki with hobbles on.