I was interested in Wikipedia when I first heard about it, and was using other wikis first. However, my serious involvement began in September last year when I discovered a situation with an article of political interest. The article had been, over a long period, carefully crafted to (1) present all the propaganda points and spin developed by an advocacy organization, and (2) to exclude all criticism.
Attempts by ordinary users to insert critical material met with immediate reversion, by certain users and by a certain IP address. Some excuse was always given. As an example, one of the "plus" arguments in the article was contradicted by a source. A user posted it with the source. It was reverted on the basis that the report was a blog. When it was pointed out that it wasn't a blog, it was an edited, on-line newspaper, it was still taken out with other arguments. The editors keeping the article free of criticism were clearly experienced with policy. When I became involved, they had to work a little harder. The IP editor crossed the 3RR bright line. I tried to file a 3RR report, but didn't get the details just the way the admin who dealt with it wanted, so even though the offenses were blatant, nothing happened immediately. But then a sock puppet account was created and *I* was reported for 3RR, though I hadn't actually done that. Fortunately, the admin who looked at it saw what was going on and blocked just about everyone in sight. The registered editors maintaining the position of the article were sock puppets of a long-banned user, and the IP address, from independent evidence, was the personal IP address of the Executive Director of the advocacy organization.
Most ordinary users did not have the patience or skills to deal with that situation.
But there is more. I discovered that an SPA had registered, more than a year before, and had been systematically AfDing articles on topics inconvenient to the agenda of the advocacy organization. Typically, nobody who understood the issues participated in the AfDs; this is a specialized field and the people who had created the articles were mostly experts or students in the field, most of them not serious Wikipedians. I.e., they don't check their watchlists regularly. Some of what was deleted was clearly not notable, but other articles were. This SPA still has an account, though his activities of late have been reduced to sporadic personal attacks. He was finally blocked for a brief period because of his canvassing for Oppose votes in my last RfA. From his activities, he appeared to be a sock puppet of an experienced user, but the most likely candidate, the same active puppet master as the one as who had been socking the article I mentioned, was not confirmed by RfCU.
This organization has funding. It has knowledgeable volunteers. I know that one of them, an experienced Wikipedian (active back to 2004, I understand) was assigned to interdict my work on Wikipedia, because he, when he realized the situation, told me.
Now, when I research the subject of the article on the internet, I find, everywhere, quotations from Wikipedia. I go to the web sites of city governments, and find quotations from the Wikipedia article that were the propaganda inserted by that organization. It's all stuff that can be asserted with a straight face; it's skillful political spin. If you look closely and check it, it is inaccurate, but in a way that is deniable, i.e., to an inexperienced eye, the difference between the statements and the truth is seemingly minor. But the difference introduces a spin, and the spin is what is important to political activists. A description of a thing becomes a "recommendation," but critical information from the same source is not mentioned. Words are shifted in definition, so that one can imply something that isn't true without actually stating something untrue. It's stuff that is familiar to anyone who has studied the tools of political propaganda. And the work of this organization on Wikipedia has been *effective.* I've slowed it down, to be sure, but it's like pushing the boulder up the hill. If there is funding and energy pulling that boulder back, it comes back.
I'm just one editor, and I have other interests. (I've become very interested in Wikipedia itself, its structure and policies and procedures, how the community functions, for better and for worse, and so, even though the article I mention above is on my watchlist, I often don't even notice the edits until much later, when I look at the article itself, so much traffic comes my way.) Wikipedia is a community project, and nothing should depend on one editor, but often it works out that way, simply because nobody else takes an interest. The article mentioned had multiple RfCs that attracted no participation. Even though it is a hot political topic in the United States.
Now, the advocacy organization I mention above is a small organization. Wikipedia has taken on a project with vast political implications. Control knowledge and you can control society. What if someone dedicates governmental-level resources to the problem of managing Wikipedia spin? Are we at all ready for this? And, since the possibility is so obvious and the possible benefit to those who would manipulate so great, what would make us imagine that it is not already happening? Just not so clumsily as with the efforts that have been discovered to date?
One person with some patience and fairly modest technical skills could develop a whole army of sock puppets, and take a few of them to admin status. But that's not the major worry. I've seen a set of articles, in another field, systematically distorted by a single SPA, active continuously for a couple of years, steading pushing the articles in a certain direction. In my paranoid moments, I realize that drug companies might have some incentive for this slant. What if some drug lobbying company paid a person to do that? It would be a drop in the bucket of their available funds. It would essentially be crazy *not* to do it. Just about impossible to prove, so, please, don't take this as an accusation, I'm just noting that such *could* be occurring. The articles have suffered greatly, whether or not this person is sincere.
Here is what I expect we could find if we had the means: we'd find a couple of corrupt administrators. We'd find more administrators who are merely dupes. The absolute numbers would be small, considering that there are, what, 1600 administrators? We would find, however, many more simple editors quietly doing their work, civil, avoiding violating policy, familiar with sourcing requirements and using them to steadily and selectively removing what they don't like and adding what they do, always with deniability. The world of "reliable source" is vast and you can find almost anything in it. Finding *balance* is much more difficult and standards for balance much more subtle.
There are solutions to the problem. But ... they fly in the face of many common assumptions, and there are obvious objections that people think of immediately. With more exposure to the ideas, those opinions would probably change, but the knee-jerk responses mean that the solutions are immediately rejected. Just as people would have rejected the idea of an "encyclopedia anyone can edit" because of certain very obvious objections. The solution is to develop true community consensus, measured within a system that allows full deliberation to take place in a connected and efficient way. It's really ideal for wikis. But we have taken certain directions contrary to it. For example, "No canvassing." If some topic comes up and is under debate, to solicit the participation of someone knowledgeable can, under some conditions, be considered canvassing. Canvassing is a problem because of participation bias, and we are totally schizophrenic about this. If votes don't count, canvassing creates a minor noise problem. But if votes count, then participation bias is a serious thing. And networks of users can exist off-wiki to do the canvassing, and, in fact, they do. This is one of them. There are others with other agendas.
Now, I claim that (1) there is a problem, and (2) there are solutions. How would we determine the truth of these? Rough consensus can be highly vulnerable to knee-jerk responses, particularly as communities become established, people become familiar with the "way things work," and the pressure to maintain status quo increases, as it always does in even diffuse "organizations" such as the Wikipedia editor community. Until we establish true deliberative mechanisms -- and we don't need to reinvent the wheel, there are many precedents for what we need to do -- we, as a community, will be a pushover for special interests. Only the clumsy ones will be discovered and stopped.