>>36
Lisp has eval built into the language. That pretty much necessitates the inclusion of a full interpreter in the runtime.
EVAL and
COMPILE are part of the language, like a lot of features, but nothing stops you from not using them or removing them from the language. If you don't need them at run time, you can safely remove them.
>>37
My only point was that while it's common for a Lisp, or other high-level languages to have a runtime, it's also possible to not have a runtime in a high-level language, just like ASM doesn't require one, however you can of course use LIBC or kernel apis or whatever from ASM, if you wish.
>>38
I don't hate the runtime. I merely wanted to prove that it's possible for mid/high-level languages to not use the runtime, if you wish to do so. I've done it in C quite a few times. Never tried to do it in CL, but I'm aware of what the runtime consists of, so I can see that for certain applications(for example a crypto lib that doesn't depend on anything), you could technically eliminate it, if you wished.
>>42
I have only been saying that LISP (for example) needs a runtime.
While it may be common to have a runtime for it to be useful, there's no absolute need to have one, and if needed, one could make a runtime-less Lisp program.
>>45
Gutting a language makes it something else, even if it still happens usable or even useful--and why would you do that just to say "look ma, no runtime!"--is it that important to you? It's not that important to me.
True, i was just trying to prove a practical point that if you wanted, a program written in a mid/high-level language could have its runtime removed if it doesn't make use of it.
There are some uses from removing the runtime in certain cases (think embedded/bootloader/small demos/shellcode/etc), but those cases are rather rare.