Path: utzoo!censor!geac!torsqnt!news-server.csri.toronto.edu!cs.utexas.edu!usc!wuarchive!udel!rochester!cornell!wayner
From: wayner@cello.cs.cornell.edu (Peter Wayner)
Newsgroups: comp.org.eff.talk
Subject: Re: Research project on hacking and viruses
Message-ID: <49614@cornell.UUCP>
Date: 11 Dec 90 21:53:29 GMT
References: <3383@skye.cs.ed.ac.uk>
Sender: nobody@cornell.UUCP
Organization: Cornell Univ. CS Dept. Ithaca NY
Lines: 37


>1. To what extent has the advent of hacking/viruses fed back into
>and affected the development of computer science? (e.g. the conceptualisation
>of genetic algorithims)

Well, I've never seen anyone say in an article on distributed
computing "As Morris states in his seminal work on viruses...", but
they really are cut from the same bolt of cloth. After the Morris
affair hit Cornell, some professors began debating whether writing a
virus was "thesis" grade material. Some felt it was "just hacking."
Regardless of this, I discussed doing a thesis on "compiler-generated
distributed systems" which, to the department, a perfectly
"acceptable" thing. 

Before the virus hit, another student was doing Certified research on
distributed computers and he would routinely clog up the network late
at night. He was about as responsible about it as he could be and still
push the outside of the performance envelope. Still, it was a pain
to find your program paging over the network when he was experimenting.

Now Robert Morris's first sin was not asking permission (either of the 
Department or of the users who's passwords he cracked). His second was
causing enough widespread damage to make the evening news. Other than
that, it was just "distributed systems research." 

I would think that his worm has taught a positive lesson to the 
community. (Although, it could be argued, that it was well-known 
before.) Namely that large systems behave quite differently than 
small ones and although he might have tested it successfully on a
small network, it crashed when it ran around the country. This 
is a very visible example of what many computer scientists were
arguing was the main problem with the Star Wars computer system. 
It could never be tested on a large scale (without a war) and so we
wouldn't really know if it would work when we needed it. 
Peter Wayner   Department of Computer Science Cornell Univ. Ithaca, NY 14850
EMail:wayner@cs.cornell.edu    Office: 607-255-9202 or 255-1008
Home: 116 Oak Ave, Ithaca, NY 14850  Phone: 607-277-6678