Primarily, our Python objects exist in volatile computer memory. They can only live as long as the Python process is running. They may not even live that long; they may only live as long as they have references in a namespace. If we want an object that outlives the Python process or namespace, we need to make it persistent.

Most operating systems offer persistent storage in the form of a filesystem. This usually includes disk drives, flash drives, or other forms of non-volatile storage. It seems like it's simply a matter of transferring bytes from the memory to a disk file.

The complexity arises because our in-memory Python objects have references to other objects. An object refers to its class. The class refers to its metaclass and any base classes. The object might be a container and refer to other objects. The in-memory version of an object is a web of references and relationships. As the memory locations are not fixed, the relationships...