Anthony DiPierro writes

I'd say Wikipedia falls short when compared to any encyclopedia that actually has a publisher.

I agree with you that Wikipedia falls far short of many encyclopedias, especially Encyclopedia Britannica. But there really are many crappy encyclopedias out there. And when Wikipedia works, it works extremely well -- I think we have a number of superb articles. If you and I differ, it is largely semantic. Wikipedia's weakness, just as a product, is that it is very uneven.

I fear that participants in the list-serve will either drop this topic entirely, or will argue over how good Wikipedia really is (or isn't) for days.

I would rather use Anthony's comment as an opportunity to have a serious discussion about why Wikipedia has more mediocre or poor articles than great ones, and how we can improve it.

Some time ago a few people - I was one, but I think Mav may have originated the idea - suggested a review board to arbitrate conflicts over content, which the ArbCom does not do. I still think this is a good idea and I think it is about time we begin a concrete discussion about what such a committee would look like, and what it's procedures would be. For starts, I would say that it could simply arbitrate content disputes by assessing the degree to which a contributor is complying with our Verifiability and No original research policies. This would be a well-defined and narrow mandate which should limit abuses. We should also discuss whether the current ArbCom or the one I am proposing should also arbitrate conflicts over NPOV.

I do not think this is an ultimate solution or a perfect idea. But I believe that no proposal will be perfect. The question is not whether this is an imperfect idea that some users can manipulate. The question is, would this help improve the quality of articles. I think it would. To those of you who think the weaknesses of this proposal outweigh its strengths, I only ask you this: please suggest something better.

And we all know that personal attacks and trolling both drain the energy of dedicated editors, undermine the quality of articles, driving away knowledgeable potential editors, and harming our general reputation. Now, I am in no position to throw stones. I have made personal attacks, and have violated the three-revert rule a few times. I'd like to think that in my case the current system worked -- other users chastised me, and I have been blocked temporarily. It cases it helped me cool down, and in most cases I apologized. I certainly think that the system works for the vast majority of contributors. It's understandable that editors who are most knowledgeable about a topic and passionate about the project, and thus emotionally invested in their contributions, may be quick to anger. But I really do believe that in the vast majority of cases, our system works.

But there are some cases where it really doesn't work. A troll can wreck havoc at Wikipedia over a long period of time, and as I said a couple of weeks ago, the larger Wikipedia gets the more territory a troll can roam over -- and the harder (certainly more time consuming) it gets to track down a troll's damage. I think that a committee to arbitrate content disputes will help; it will take some pressure off of the ArbCom, for one thing. But I don't think such a committee will be enough to deal with serious trolls. We need a more efficient system for banning trolls.

I bet many of you are thinking, Steve's definition of a troll is simply going to reflect his own bias or agenda. Who is to define what a "troll" is (in other words, who is going to enumerate the kinds of, or measure the amount of, personal attacks that distinguishes a troll form the many well-intentioned users who just lose their temper some time? Obviously the answer is, "not I." No one person can do this. But enough of us have had enough experience that I am sure if we put our heads together and worked at it, we can come up with something appropriate and effective. I think if we revisit our No personal attacks policy, we can draw on a lot of experience to improve it - to make it systematic and specific enough that there will be little argument over what problems ought to go to the ArbCom and what cases can be handled by admins. This means giving admins clearer guidelines as to what kinds of personal attacks (or frequency?) are grounds for a short term block, a medium term block, or banning.

I have forwarded a few concrete proposals. None are perfect, all are undeveloped. My hope is that they provide a basis for a constructive and practical discussion that will lead to better and more fully developed proposals.

I've expressed many times how much I value the anarchic elements of Wikipedia, that it is an open community and relatively transparent. But as several people have occasionally had to point out, this process is the means to an end: a high-quality encyclopedia. Anthony is right that we fall short. We need some more structure, or new processes, that will address our weaknesses. I am certain that we can do this in a way that does not undermine our principal strength -- the openness of our community -- but rather in a way that supports this open community by providing more effective processes for constructive collaboration and productive debate that is as respectful as it is impassioned.

The question: where do we go from here?

Steve

Steven L. Rubenstein Associate Professor Department of Sociology and Anthropology Bentley Annex Ohio University Athens, Ohio 45701