But then you start adding to it. Soon you find yourself looking up all the odd new commands. And wishing for a WYSIWYG editor because you can't remember the commands or not sure what it will look like without the live render.
It's a bit like saying, "Hey this QWERTY keyboard is nice, but what if it had keys for all the Cyrillic, Devanagari, Chinese, and Arabic characters too? Wouldn't that be great?" Well, yea. But you just put the hunt back in hunt and peck.
By that I mean, the basics reuse the way we faked formatting to do real formatting. The input is (usually) perfectly readable on its own.
Even if you don't know (or remember) how to author a markdown file, you can probably still read it just fine. The tables still look like tables. The paragraphs are just paragraphs.
I do still have to look up how to do stuff in markdown sometimes. And that's fine. Your active vocabulary is always smaller than your passive one.
So the way I judge this is by how readable the input is.
I'm not sure how well they succeeded at that. A lot of what they show doesn't really add to or take away from that.
But I didn't see any examples of them formatting math. I only rarely use LaTex. And when I do, it's not because I need a "paged" mode or need to include an author. It's because I need to format something markdown can't do, and that's usually a math equation.
So I am curious how that ends up looking.