The goals of some peers need not be the goals of the designers or the other 
peers.  A real-world example: some peers want to share files, and other peers 
want to prevent any peers from sharing those files.  The self-interest of one 
peer looks like malice to another peer.

A peer that is merely self-interested from its own point of view may thus 
easily be malicious from the point of view of the designer and other peers.  
For a designer or peer to assume mere "greed" is thus to introduce unknowable 
and uncontrollable assumptions about other peers' motivations.  That is why 
maliciousness, and not mere greed, is the normal threat model used in computer 
security.  Such worst-case behavior is even a common assumption in reliability 
analysis, where it is called "Byzantine" behavior.

> On Jan 2, 2008, at 10:07 PM, [EMAIL PROTECTED] wrote:
> 
> > What would be the peer's benefit from lying or dropping  
> > certificates?  I agree this is a possible attack vector but from  
> > the perspective of a greedy, but not malicious peer there's nothing  
> > to be gained from such behavior.

_______________________________________________
p2p-hackers mailing list
[email protected]
http://lists.zooko.com/mailman/listinfo/p2p-hackers

Reply via email to