A lot of Python's complexity like descriptors and metaclasses and the so-called cooperative multiple inheritance come from an era where object orientation is gospel and yet people keep finding limitations to what can be accomplished in an OO paradigm and therefore they patch up the language to support ever more complicated use cases.
I have a feeling that if metaclasses and descriptors were to be proposed today it would be laughed out of the PEP process completely.
> I have a feeling that if metaclasses and descriptors were to be proposed today it would be laughed out of the PEP process completely.
I think you need at least one of these. Modern Python eschews metaclasses but descriptors are everywhere. Without them you couldn’t have @classmethod, @staticmethod nor @property - or at least, you couldn’t generalise custom method lookup and all three of those would need to be special cases.
The alternative approach is to drop descriptors and make method lookup much simpler, but then you need metaclasses (see: static methods in Ruby).
If descriptors didn't exist in the language, classmethod, staticmethod, and property would become keywords. Languages like Java and C++ already have static as a keyword; C++ also made its type system have a special pointer to member type; such behavior can be hardcoded without having a flexible descriptor mechanism.
Custom method lookup would just be the same as custom attributes, that is, by the user overriding __getattr__.
I have a feeling that if metaclasses and descriptors were to be proposed today it would be laughed out of the PEP process completely.