Matus Tejiscak wrote:
zygohistomorphic prepromorphisms
Please tell me this isn't a real technical term. o_O As for concrete suggestions... I've always thought we could do more to use static information about the program to aid runtime GC. It's no deep secret that destructive updates are essentially like a compile-time / coding-time GC operation. You determine before runtime that the old version of the data will never be needed again, and hence update it in-place. Making this kind of thing more automatic could be interesting theoretically and practically. The other thing is connectedness; a GC performs a sweep of a big chunk of memory to find live objects, but if you somehow knew from compile-time analysis something about what the runtime linking structure is going to be, you might be able to do something interesting. (OTOH, laziness and sharing probably spoils this.)