🚀 go-pugleaf

kludge@grissom.larc.nasa.gov (Scott Dorsey) writes:

> Viruses aren't a real problem in any system that has memory protection and
> any way to prevent users from directly writing into system space.  In such
> a system, a user would be able to destroy his own area, but wouldn't be
> able to touch the OS.

While your second statement is, of course, true, the first is not
necessarily so. Viruses are a problem is there is sharing,
functionality, and transitivity of the information flow. If you
prevent any of those, you make the viruses impossible. The only
problem is that you also make the computer not very useful.

Let's consider Unix - a typical system with memory protection,
discretionary access control, and all that jazz. Absolutely nothing in
this system prevents a virus from infecting ALL executable files of a
particular user. Between the different users on one and the same
system the virus will not spread that well, because users don't often
execute programs that belong to other users. An exception is the
"users" bin and root. Once one of those users executes an infected
file, the virus will spread in the system like a wildfire. But again,
those users do not often execute programs that belong to other
users...

Between different systems the virus will have even less chance to
spread, because the sharing of programs occurs mainly in the form of
source code. You do not walk around with infected tapes and do not run
the programs on them on different machines, as you do in the PC world.
That's why, in multi-user systems the worms and the trojans are of
much more concern than the usual viruses.

Memory protection has almost nothing to do with it - ok, without it
the viruses will be easy, just as in MS-DOS, but even if it is
present, this does not prevent the viruses from spreading. What
prevents them is the limited sharing.

Regards,
Vesselin
-- 
Vesselin Vladimirov Bontchev          Virus Test Center, University of Hamburg
Tel.:+49-40-54715-224, Fax: +49-40-54715-226      Fachbereich Informatik - AGN
< PGP 2.1 public key available on request. > Vogt-Koelln-Strasse 30, rm. 107 C
e-mail: bontchev@fbihh.informatik.uni-hamburg.de    D-2000 Hamburg 54, Germany