Talk and discussion are completely different process than collaborative editing, and they demand different tools.
I also agree about the markup (it's markUP. "Markdown" is a specific type of markup language which, despite its flaws, probably is the easiest and now most common one)
In this case, "hey we need a way for people to talk to each other", "sure, we should make a discussion tool, but to get it right now we could use a wiki page that users edit one after another to add their answers".
The fact that it started as such is not an issue per se, but after so many years wikipedia should have replaced it with something better and easier.
Lean startup methodology was an inheritor of common wisdom such as "worse is better" and "best is the enemy of the good" but they're all in the same family tree.
I don't see MediaWiki markup as byzantine or archaic, except for whatever half-assed LaTeX processor they use. It seems to only recognize some random subset of features from various ams packages. Why they don't just use a fully-featured TeX distro is beyond me.
Wikipedia's talk page workflow (indenting to start a new comment, then inserting a special image to wrap around to the margin once too many indents have accumulated) is moronic.
I've contributed edits to maybe a dozen or so different Wikipedia articles, and would like to contribute more, but it's hard to get past a couple of hurdles.
The main one is the byzantine and archaic editing tools. The markdown language that Wikipedia uses is a PITA to learn, and following a discussion on the talk page is almost impossible. Wikipedia presents a wonderful UX for browsing and reading. It has perhaps the worst UX I've ever seen for a beginner who wants to make a contribution.
Beyond that, I spent a significant amount of time editing some articles on a fringe religious group that I had studied extensively. (Apart from the group's own leaders, I arguably knew more about the group than anyone else in the world.) But the group that was the focus of the article kept backing out my edits and replacing them with propaganda. It was simply too much work to fight them (especially with the horrible edit/discussion tools), they were a lot more motivated than I was, and I eventually gave up.