I'm not a huge fan of Wikipedia, perhaps for the reasons one might expect. The idea is somewhat sound and parts of it are useful, but their attempts to establish “academic credibility” are ill-considered. And as long as they insist on having lengthy articles about people and events they'll eventually end up turning into just another encyclopedia.
The most sensible thing is to never use it, and for a totally obvious reason: it's not reliable, and in its present form never can be.
While no encyclopedia is totally objective and reliable, most encyclopedias are not maliciously unreliable. Encyclopedia editors strive for maximum accuracy, and will be financially liable for gross deliberate inaccuracies: bad advertising, lost sales, lawsuits, and, ultimately, bankruptcy.
At Wikipedia no one is truly responsible for content or accuracy, and in theory no one at Wikipedia will ever be financially liable for any lack thereof. There's no defense against deliberate vandalism. Decisions are made by whoever yells loudest and is most adept at manipulating the Wikipedia community, not by the most qualified.
In terms of outright vandalism, they're fortunate no one has really decided to be nasty. With a zombie PC army (a Weapon of Mass Defacement?) it would be trivial to deface every article. Although these changes would be detected, someone would have to revert the changes... and these changes would keep reappearing with no way to block them. (They had a similar problem a couple of years ago, but that person was only using one IP!)
Worse, as the project grows it will take increasingly longer to react to such problems. Layers upon layers of useless management are building up.
The Wikipedia premise is that most people are good-natured... but what most people are has nothing to do with it. It only takes one ill-intentioned (or single-minded) person to make a mess of a thousand people's efforts. Most of us have seen that happen before, many times.
Many Wikipedia editors keep approaching (and then immediately shying away from) what will be the final answer: true article maintainership, where someone has to approve each change. This will invariably impose bias. The bias problem already exists in a more subtle form, however, and it's better to make it blatant. Most articles are slanted towards the beliefs of the most active contributors, as their words carry the most weight and they have the most time to repeatedly redact articles.
OSS projects work because they're usually controlled by a few people (preferably one) and are objective in that the software either performs as advertised, or doesn't. Free-for-all OSS projects end up being infested with trojan horses, bugs, and stop working very quickly.
Wikipedia is an OSS project, only it's text instead of code and thus isn't objectively testable. To grow beyond a certain level of quality it desperately needs dedicated maintainers. (A head honcho to decide which articles are kept, and individual people for individual articles.)
If you don't have this you end up with exactly the problems they face today. Committees can't do it, a group of more than two people can't make effective consistent decisions (and the results are rarely better than if just one person were doing it anyway).
Wikipedia's strength, and where they should focus their efforts, is the ability for people to write about obscure factual subjects. There's plenty of useful information which isn't always written down or accessible in obvious ways, particularly histories of things and specialist minutiae. Putting this material in a searchable online encyclopedia is useful, and adds tremendous value to Wikipedia as compared to a mainstream encyclopedia or the Web at large.
The idea of having useful articles about controversial ideas is laughable given that anonymous editing is permitted. As one might expect, none of these Wikipedia articles manage to stick to the facts. (This includes anything mainstream; it's eye-opening to see what qualifies as controversy in some people's eyes.)
Another major impediment to a truly neutral Wikipedia article is their overall approach to “NPOV.” The creators believe this means you have to espouse every viewpoint, even ones which have no basis in fact whatsoever. If more than three people believe something, in it goes!
Finally, in some sense Wikipedia is missing the point. If I want to know what's going on right now with panda bears, I'll be better off using Google to look for people doing active research and maintaining web pages about panda bears. Since users have to verify anything on Wikipedia against other sources anyway, they might as well go to the other sources in the first place.