On (26 Oct 97) Marius Bendiksen wrote to Jerry Coffin...
 MB> > Yes, it does.  I/O under a DOS extender is measureably slower than
 MB> under > DOS.
 MB> Actually not much, unless the DOS-extender is of the sort which
 MB> switches back to real mode, or uses normal V86 mode to use the BIOS
 MB> interrupts.
Hmm...are you implying that there is another sort of DOS extender?  If
you don't switch back to V86/Real mode, you can't use DOS, and if you
don't use DOS to do your I/O, you don't have a DOS extender anymore...
 MB> > Sure they have.  Microsoft and IBM did one a long time ago called
 MB> > OS/2. Since then, Microsoft has done another called Windows NT.  A
 MB> > few years ago a guy named Linus Torvalds did yet another that's
 MB> > generally known as Linux.
 MB> Based on the BSD 4.4 Lite kernel, the FreeBSD project also did this.
True - I wasn't trying to give an exhaustive list by any means.  There
are also several more versions of UNIX and similar systems such as Hurd
(yes, after only 14 or so years of arguing, it's actually in beta
testing) SCO, Interactive, Coherent, the 386 version of Minix, and so
on.
 MB> Quite true, but consider the fact that many people /have/ already
 MB> recieved the Nobel prize. Maybe it's time for somebody from 80xxx to
 MB> be next? :)
Maybe so.  Anybody who wants to try it is welcome to go right to work on
the project.  OTOH, the Nobel committee might not agree with me as to
the value of the work involved... 
    Later,
    Jerry.
--- PPoint 2.02
---------------