Måns Rullgård <[EMAIL PROTECTED]> wrote: > Raul Miller <[EMAIL PROTECTED]> writes: > > > On Sat, Jan 22, 2005 at 09:58:00AM +0100, Måns Rullgård wrote: > >> Interpreters are a different issue from the exec() situation. The > >> program being interpreted generally does not communicate with the > >> interpreter at all. > > > > If the interpreted program and the interpreter can't communicate, then > > usually nothing works. Variable values are unknown, control flow never > > happens, and so on. > > The interpreted program interacts (I don't think "communicate" is the > appropriate word) with the virtual machine (in a loose sense of the > word) presented by the interpreter. It does not communicate with the > actual implementation.
This does not make any sense. Of course it communicates with the actual implementation. It does so through the virtual machine. It seems like you are saying that when I send this email, I am not interacting with you, but only interacting with the keyboard. > A regular program interacts with the registers, memory and so on > found in the machine, not with the individual gates, electrons and > whatnot that make up the actual hardware. So if I write a program that puts things on the display and reads keyboard input, that is not interacting with the physical device? Regards, Walter Landry [EMAIL PROTECTED]
