On Tue, Sep 29, 2009 at 09:15:34AM +0200, Dan Kaminsky wrote: > Infections by these rare payloads would constitute a sort of "long > tail" of malware -- too rare for a signature, but in aggregate, > possibly common enough to represent a significant number of > infections. > > But how common? I mean, we know the long tail doesn't work exactly as > promised in the media space. We also know there's a lot of infected > boxes out there running AV. It'd be really interesting if we had data > around this question.
This is a fascinating question. And there's certainly precedent for abusers to operate in this fashion: consider snowshoe spammers, who distribute their presence and their activities widely in order to minimize the observables, thus decreasing the risk of detection. Given that and other similar tactics, it wouldn't surprise me at all to find that distribution-limited malware has been deployed, in an attempt (again) to decrease the risk of detection, and thus to forestall countermeasures by vendors. But I must admit that, at the moment, I'm at a loss for a methodology by which we could approach this question in a meaningful way -- that is, a methodology that would quantify the answer. ---Rsk _______________________________________________ Fun and Misc security discussion for OT posts. https://linuxbox.org/cgi-bin/mailman/listinfo/funsec Note: funsec is a public and open mailing list.
