Subject: Re: Cons Cell Representation---`sameness' again From: Erik Naggum <erik@naggum.no> Date: 1999/04/08 Newsgroups: comp.lang.lisp Message-ID: <3132551629497789@naggum.no> * Barry Margolin <barmar@bbnplanet.com> | You've never heard of a machine-level debugger? Get yourself a Lisp | Machine and type (si:ddt). I also recall low-level functions that would | return the address that a locative refers to, as an integer. yes, of course I have heard of them, I just find it wholly irrelevant to a discussion of object sameness, which I tried really hard to show with the virtual memory page example, which goes below visible machine address to physical machine address. | Or if you have a foreign function interface, the foreign language may be | able to convert the address of a Lisp object that you pass to an integer, | and return that integer. yep, that's the violation of the sanity of the system. | This is what was meant by a machine-level microscope. It's operating | below the level of the Lisp language, looking at the hardware | representations. but, but, this microscope has memory and shows you what was before in a way that is entirely artificial. I mean, copying GC is just like paging the same data in at a different physical memory location. in effect, this "argument" drops an important context: instead of saying "the bit strings" it should say "the bit strings at time T1" and "the bit strings at time T2". but that would make the silliness explicit... | I agree that this is not really a very useful level to mention in this | thread. whew! :) #:Erik