[virtmach] Definition a virtual machine ?

Laurent Martelli martelli@iie.cnam.fr
17 Nov 1999 17:29:07 +0100


>>>>> "Ian" == Ian Piumarta <Ian.Piumarta@inria.fr> writes:

  >> How would you define a virtual machine ?

        [...]

  Ian> I don't think it's possible (or even realistic) to try to
  Ian> define a virtual machine in terms of what it should or
  Ian> shouldn't have (complete instruction set in software, a virtual
  Ian> memory implementation, whatever...), since wherever you draw
  Ian> the line you'll be making someone unhappy.  The best we can do
  Ian> (and I think someone already hinted at this) is to say that a
  Ian> VM is *anything* that extends the programmer's view from that
  Ian> of running the bare hardware -- a definition which subsumes
  Ian> both the "OS" and the "language" interpretations of "virtual".
  Ian> And like someone else already said, the existence of a rich,
  Ian> complete virtual instruction set -- bytecoded or otherwise --
  Ian> is just an implementation detail related to where one chose to
  Ian> draw the line between abstract and concrete.

  Ian> Does any of that sound even vaguely reasonable?

I think it is. I believe it is mainly a human's language issue in
fact. Some people use the same words with slightly or completely
different meanings all the time. I just thought than since this is
supposed to be a ML about virtual machines, we ought to have a common
definition of the word so that we can better communicate and avoid
missundertandings. 

-- 
Laurent Martelli
martelli@iie.cnam.fr