At issue here is that many of the things they choose to pour their delighting syrup over are fundamental to the way the project is managed. We probably don't need to litigate whether MediaWiki's UX could be improved; it can. "How?" is the operative question; one clear answer is probably "Not this way."
That's the problem with designers, they always want to "enhance X with great design", ignorant of the fact that the object of their OCD desire for "great design" (in this case the world's most popular web service) obviously has great design already, as it's the most popular website in the world.
This is a horrible blanket statement. Don't call all designers ignorant because you don't like one specific proposed redesign.
> obviously has great design already, as it's the most popular website in the world.
Amazon has (sorry for anyone who works there) hideous design. HIDEOUS. It is a very popular website.
There are dozens of examples of websites with bad design that are popular. Many web-forums have thousands, tens of thousands, of users yet are really hard to use.
It's not particularly hard to use. You think of a topic you want to read, type it in the search and press enter. You start reading that topic. And you can click on a hyperlink to related articles within that article.
Great design already, with a few rough edges. Nothing that needs a fundamental overhaul. I would hate to see Wikipedia go the way of Digg!
These guys did that with this brief. It's not the right answer, but it does raise some interesting questions. How do you design a more delightful Wikipedia without damaging the content? Because surely you agree that certain elements of Wikipedia - like editing pages, for instance - could use a good bit of delight here and there.