Path: utzoo!mnetor!uunet!mcvax!ukc!dcl-cs!bath63!pes
From: pes@ux63.bath.ac.uk (Smee)
Newsgroups: comp.sys.atari.st
Subject: Re: Virus--  suggestion!
Message-ID: <2377@bath63.ux63.bath.ac.uk>
Date: 25 Mar 88 18:55:28 GMT
References: <1288@uop.edu>
Reply-To: pes@ux63.bath.ac.uk (Smee)
Organization: AUCC c/o University of Bath
Lines: 40
Keywords: virus, GEMDOS, atari

In article <1288@uop.edu> exodus@uop.edu (EXODUS) writes:
>
>What if someone wrote a 'friendly' virus.  One that fixed the bugs in the
>Operating System.  One that multiplied so everyone's bugs were fixed. 

No, no, a million times no.  There is NO SUCH THING as a 'friendly' virus.
Consider the original Amiga virus.  It wasn't meant to have any ill effects.
It was meant simply to sit around for a while, and then to pop up and say
(basically) 'hey look, guys, I'm a clever person who wrote a virus.'  And,
indeed, that's all it does.

EXCEPT that if it finds a write-enabled disk, and installs itself, and the
disk happens to be a (probably expensive) copy-protected program, even that
is deadly.  UK dealers alone have lost thousands of pounds worth of stock
to friendly viruses -- and the UK is hardly the world's biggest micro
market.

The other problem is that it would (unavoidably) interact (at best) or
interfere (at worst) with programs which attach themselves into the system
vector chains for specific known purposes -- and there are a lot of handy
programs which do that.

Even if you could avoid *that* (which I doubt) any resident program is going
to screw up someone's (probably my) very delicate space-allocation juggling
act.  I've got ramdisks, spoolers, and ACCs very delicately balanced on the
various disks I might boot from, to leave just enough room for the major
application I might run in that environment.  A few odd K sucked up for
something else would probably break many of my operating environments, and
then I'd have to go thru finding and de-virusing at least the critical
disks -- no matter HOW friendly the virus was.

I could get into the idea IF it could be demonstrably shown that the losses
would be small, and the gains large -- but I don't believe you could show
that (or guarantee it had been done right).

If you want to write a bug-fixer which an owner can install at will on
disks of his or her own choice, and which you believe will provide fixes
for all the bugs, fine.  But you've got no business running programs on
other people's machines without them being aware you are doing it.  Period.
And that's what your suggestion amounts to.