The "orthodox" comes from a specific type of GUI, namely one that is driven by commands under the hood. UI elements are merely used to trigger commands that have the actual effect, and these commands could just as well be executed by hand, or automated into more complex commands.

This is an excellent way to build powerful UIs. It is what drives things like Vim, and often why Lisp-based software is so hackable -- think Emacs, StumpWM, etc. Instead of writing plugins against some small plugin API, you're wiring new functionality directly into the application.

The article you reference goes into more detail, as you say.

Does Blender also qualify? It even shows you the name of the Python function behind each UI element on hover, which is great for discoverability when scripting. Or maybe it used to, can't see it now.

Similar to the ELM architecture in a way? Except that the commands are literal commands that can be executed outside of the standard UI interaction.

Isn't this what tools like lazygit use?