Srdjan Todorovic todorovic.s@googlemail.com wrote:
On Sun, 18 May 2008, MJ Ray wrote:
I think we're at much more risk from trojans (binaries from tainted sources), worms (programs that break-in and replicate - Stoll later realises that the virus above is actually a worm) and crackers.
True, but just because a binary is a trojan, does not mean it can't also be a virus and a worm. They're not mutually exclusive. So thus your "we're at much more risk" statement is not always going to be strictly true, as you could have a hybrid malware application.
Not being mutually exclusive does not mean we're not at much more risk from trojans, worms and crackers. I meant to express the relationship mean(risk(virus)) << mean(risk(foreach({trojan, worm, cracker}))) and I think that's pretty surely true, even in light of a hybrid malware application, because that's going to increase risk(virus) but also increase risk on the other side too.
Hope that explains,