If you interpret the contents of a "plain text file" through anything more complex than a Unicode text encoding, then you now have "a special format". (And if you don't, you still at least need some encoding, even if it's an implicit ASCII encoding. "There ain't no such thing as plain text", as they say.)
There are also plenty of "special formats" out there that are really just zipping up a directory tree of something simpler.
What urtext - to someone who moved from org-mode to Obsidian - seems to do is
* chain you to a proprietary text editor that does some python interpreting. I wonder which use cases would want me to have my document change itself (or even change its own changing logic).
* highlights 'features' that really do not live in the document itself but rather in the editor's logic (like timestamp handling)
* introduces a complex structure to express 'nodes', which appear to be essentially text anchors.
But anything else, the editing, the easy-UI-free syntax, markdown has done ten years ago.
So it is not a text format, because it lives in a very specific editor.
It is not a fully-fledged software package either.
It sounds a lot like some sort of macro language which woke up one day and decided to rather be a text repository.
If you (speaking of a general you, not you specifically) wanted to convince people to make the switch, a list of barely described features is insufficient. You need to sell a solution, not a product.
Your answer contributes nothing, which is just shy of what the intro pages offers in relation the question. Everything urtext claims, is readily achievable through markdown + some collection of tools of your choice, for which many exist, some being all in one solutions. Obsidian with a few community plugins is a clear example.
I enjoy and appreciate projects like this, I _WANT_ to like and understand it, but as it stands the information simply isn't there.
There are also plenty of "special formats" out there that are really just zipping up a directory tree of something simpler.