Reference counting garbage collection

[Paul Rubin]

Is there any particular reason Python uses reference counting garbage
collection (which leaks memory if there's circular structure) instead
of Lisp-style garbage collection?  As Python gets used for more large
application development, it gets troublesome to make the programmer
worry about whether the data has cycles.  Also, using a compacting GC
helps localize memory references, speeding up the program by improving
cache performance.  As CPU's get faster and memory doesn't, the cycle
cost of a cache miss continues to increase.  So maybe the GC
implementation should be revisited at some point.