Talk:University of Canberra/RCC2011/What key facilitation capacities help create successful wikis?

Is Wikipedia a "successful wiki?"

It's clear that Wikipedia has been successful in certain ways. If I want to look up the meaning and some explanation for some scientific term, usually I get a decent explanation. But I've discovered that if I'm familiar with a topic, and happen to look, I can often see that the article is quite defective, sometimes seriously. I used to think, "Just fix it!" However, sometimes -- certainly not always -- there will be editors who will resist the fix, sometimes based on technicalities, but sometimes obviously based on either some point of view, or a kind of default point of view: "I'm right and it looks like you disagree with me, so you are wrong." If this editor has friends who are administrators, forget about it. It's impossible, unless you also have those friends or are expert in using the structure and willing to put in the considerable effort it can take. Just to make one little change to an article. Horribly inefficient. And you can go through all this, and look at the article a year later, and it backslid. w:Sisyphus.

Whenever there is real-world controversy over a topic, Wikipedia can be grossly unreliable. The basic mission was to create a compendium of "all human knowledge," but that was an overstatement. Down in the nuts and bolts, "all" meant "all notable," which, in the end, by policy -- but not necessarily in practice -- meant "all knowledge covered in reliable sources so that it's verifiable." If you know a thing, your friends know a thing, and you are the world's experts on a thing, it means nothing, though it certainly is human knowledge. Yet, Wikipedia is full of unverified information, because people believe the advertising: all human knowledge, and no practical structure has been developed to ensure that information actually is verified, and, perhaps even more important, that coverage is neutral and balanced.

Another fundamental aspect of the mission was "neutrality." Wikipedia articles are supposed to adhere to NPOV, or "Neutral Point of View." Great idea. It's subjective, however. "Neutral" is a relative term. It's easy for cherry-picked text to appear neutral to those who don't know the subject. It's easy for sources to be quoted or used out-of-context, so that what the text says seems verifiable, but any subject expert would see the problem in a flash. Someone who favors a point of view may see material that confirms that point of view as "neutral," and text that might seem to contradict the POV is, of course, "POV." Even if equally verifiable. And, indeed, it might be. But it might be needed for balance, properly framed and attributed.

One of the common phenomena, in articles under dispute, is "cherry-picking." Evidence is picked and inserted that appears to support one point of view, and if one looks at the editor's contributions, a curious phenomenon can be seen. Even with a complex controversy, with mounds of evidence on both sides (and sometimes there are more than two sides!), the editor is always inserting or restoring evidence on one side, and never inserts "inconvenient evidence." Does this mean that the editor is biased? Almost certainly. However, is there something wrong with this? Not necessarily. It depends on how the editor goes about it.

There is only one way to approach neutrality, and that is to maximize true consensus. Jimbo Wales understood the principle; if text is truly neutral, all sides will see that their point of view is fairly represented. Factions only get exercised when their point of view is unfairly denigrated, but with proper balance and completion of coverage, according to sourcing standards, all factions should be able to sign on. There can still be "die-hards" who won't accept the insertion of "inconvenient fact," rather than accepting it with due balance, but these are, in my experience, rare, and they have generally been trained to be intransigent by experience with the success of intransigence on Wikipedia. That happens when the user's point of view is MPOV, i.e., aligns with the "majority point of view," which means "majority among Wikipedia editors, especially administrators," not necessarily "majority" as in "majority of experts."

Wikipedia did not develop, and apply widely, decent consensus process, designed to maximize consensus. Instead, the project went toward maximizing the number of articles and the bulking up of content. Consensus process is generally time-consuming. It requires detailed discussion. Wikipedia is famous for harassing and banning experts, because experts typically love their subject, as well as knowing about it, and they will often assume that other editors will be similarly interested. They are called "SPAs," single purpose accounts, and are given short shrift if there is a conflict. An editor with wide contributions and an editor who is, say, a professor in a field, who only contributes in that field, are no match in a conflict. As the administrative structure consists of users who aren't SPAs, their sympathies are naturally with the wide-focus editor, the "content contributor," and surely banning this editor who only works on one article won't be a big loss!

I used consensus process in a few cases I came across of conflict between an expert and a general-purpose editor, and successfully. It's not that difficult. But as soon as I became expert in a field myself, I was practically dead meat, and in my field of (new) expertise, the Wikipedia article is atrocious, heavily biased to what became the "scientific consensus" twenty years ago, but which, in recent years, flipped. In spite of recent peer-reviewed literature being 100% in one direction, and that is hundreds of papers published in mainstream journals, including 19 positive reviews of the field, over the last six years, the Wikipedia article excludes almost all the recent review and work, and often presents the topic with radical misinformation, clearly false -- which can be sourced from newspapers and other popular sources -- and because experts have been banned from the article (at least three), and non-experts who were aware of the current science were marginalized, edits reverted consistently, the article bias continues, and has been, in fact, getting worse.

When administrators intervene and see an expert as "biased, POV-pushing," they tend to warp the process. Experts are, in fact, biased. They tend to believe that they know the field they have studied more than those who have not, or who have only looked at cherry-picked sources. Whenever consensus process was followed on Wikipedia, with respect to material in this field, article content improved. But whenever experts were banned, and those who had tried to negotiate consensus were banned -- because they were fostering a lot of discussion, which is somehow considered disruptive -- article content declined.

Am I biased about the field? Perhaps. I started out with the general opinion, the "skeptical position." But then I read the sources, and I actually bought the books (on all sides of the controversy) and, it turns out, those who read the sources tend to have a different perspective than those who don't read them, who assume that anything that contradicts their opinions must be bogus, that peer-reviewers have been duped, that thousands of scientists are deluded, wishful thinking. Skeptics in the field have been unable to be published under peer review in relevant mainstream journals for at least five years. ("Skeptic" here means those who reject the entire field, not those who are normally skeptical about this or that result.)

Now, I know how to go to ArbComm, and I was successful going to ArbComm over the two issues I took there. But ArbComm isn't sophisticated, and tends to try to protect the majority. I.e., if there are two dozen editors yelling at one (me!), and even if ArbComm agrees with the point of the filing (what I was asking for), and implement it, they are worried that they will offend the two dozen, who include "valuable contributors." And, at the time of my last case, they had not clearly idenfied the two dozen as a "cabal," a faction that operated more or less coherently to exclude certain POV and reinforce their own. They did later, and that particular faction has been "defanged," more or less, but the operating principles haven't changed. ArbComm can, and will, shoot the messenger, and only a minority of arbitrators are inclined to stop it.

So I could go to ArbComm, but doing this in the presence of an opposed faction is quite time-consuming, and, eventually I concluded, just too much work. A single article isn't worth it! Now, repeat this over thousands of articles on controversial topics, with different experts or informed and interested users....

Wikipedia is failing at its basic principle of neutrality, because inadequate consensus structure was built, and such structure threatens the existing oligarchy, so it can't be built there. It's still possible on Wikiverity, I believe, or I wouldn't be active here. --Abd 16:35, 4 February 2011 (UTC)