Ironically, the original poster suffers from having looked at a particular article at a particularly bad timeslice and gotten an ugly result. The present text of the Bill Gates article is greatly improved. Interestingly, this is exactly the sort of problem that my wikipedia talk looks toward addressing.
Furthermore, its important to understand what wikipedia is and what it is not. Wikipedia is not a replacement for a traditional encyclopedia. This does not mean it isn't useful. A famous engineer's cynicism is: Cost, Speed, or Quality, pick one. An Encyclopedia is a model that picks Quality. Encyclopedias are slow and expensive, but the results are good. Wikipedias are fast and cheap, and the results are not as good.
If you want to teach 11 year olds about the history of Greece, you don't want wikipedia. They may get bad information, they can't easily reference a particular revision (most people don't understand how to do that with wikipedia), and they are going to be exposed to poor grammar and poor structure at a time when you are trying to teach them how to communicate effectively.
If you want to learn about a terrorist incident that occured two months ago, an encyclopedia is of no use. You could turn to the press, but old press articles are hard to find, and Wikipedia is often a vastly more useful resource, because it presents information in a matter of fact way and often draws from a wider array of resources (including press reports which form a primary source material).
Wikipedia fills the gap between the bleeding edge of the headlines and the cast in stone of dusty reference materials in a way that no other resource can. The sooner people realise that every tool doesn't have to solve every problem the better they'll be at figuring out how to make their tools really succeed at the particular things they are well suited for.
it suddenly occured to me that as there needs to be a way of judging, other than the facility of editing it, the quality of ( or usefulness, or a variety of criteria) of articles on wikipedia.
how about a simple facility of voting for articles and obviously articles which consistantly score badly should pop up for review.
although should each edit deserve a fresh score or should past editions be included but weighted according to the edit
that way adding a comma won't remove a particularly good set of scores and thus discourage edits or adding a comma set to zero a deservely bad score.
plus a scoring system is in accordance with wikipedia's democratic philosophy and adds an element of meritocracy.
systems need feedback
plus different writers could get, like the reputation agent, different scores
