Fun fact: There's a pretty compelling case to be made that early photosynthesizing organisms were purple instead of green, so that for about half a billion years the earth was purple and blue like a cyberpunk street scene.
This is not standard practice. An article that is controversial or one that has been vandalised a lot may put in place such a policy. But the vast majority of articles on (English) Wikipedia can be freely edited.
I'm not talking about updating an existing article. I'm taking about publishing an entirely new page. I can and do make corrections and additions to existing articles without review. I wrote a completely new article and every time I submit it for review it comes back with a different reason for rejection. However, the most recent one was actually due to a misunderstanding on my part about acceptable sources (turns out I was being more restrictive than I needed to be), so at least it'll be easy to implement the changes this reviewer wants to see.
Nope! Just some technical stuff. Maybe it's one of of those things where there's not technically an enforcement mechanism. I read all about how to start a draft and turn it into a full article, which includes submitting it for review. Maybe you can just decide not to do that.
I'm guessing that's a new guideline then. It wasn't there when I joined. Also I might have been granted autopatrolled at some point, which might be why I get away with it.