Improper analysis leads to improper conclusion [LWN.net]

Improper analysis leads to improper conclusion

Posted Jun 17, 2006 22:50 UTC (Sat) by Max.Hyre (subscriber, #1054)
Parent article: PCs to developing world 'fuel malware' (Register)

His argument for the increase in malware is based on two premises:

Crime levels are determined by the following factors:

the size of the community (the bigger it is, the higher the number of potential and actual criminals)
the level of economic development (it's easier to earn a living by honest means in more developed countries)
the ability of law enforcement bodies (e.g. the police) to investigate crimes and imprison the perpetrators

and

A particular cause for concern is programs which advocate `cheap computers for poor third world countries') - these further encourage criminal activity on the Internet. Statistics [...] confirm this: the world leader in virus writing is China, followed by Latin America, with Russia and Eastern European countries not far behind.

There's a subtle distinction between the subjects of the these two paragraphs which renders the conclusion untenable: the first discusses the integral community, while the second is about individual sections of a community. The crime rate in New York City is a result of the first set of items, while the crime rate of Times Square is a result of the amount of money for the taking (to which the number of thieves is proportional), the ability of criminals to get there, and the number of police officers stationed there. None of those have any more than a tenuous connection to the first set of factors.

Thus, the since the amount of money available (everything in the world, to a first approximation :-/) and the ability to enforce the law are unaffected by the distribution of laptops, the question is what is the effect on the second factor (economic development). I can't see OLPC making things worse, and quite likely improving things (see hingo's comment), I see no justification for Kaspersky's conclusion.

Kaspersky's article is mainly about the anti-virus market, and mentions Microsoft's getting into the field...

<paranoia>
It must have occurred to someone in Microsoft that there's a nice market in building vulnerabilities into their applications so their anti-virus can be first, and most effective, against said vulnerabilities. Of course, it couldn't be done blatantly, but still.... They certainly don't seem to be in a hurry to detect their business partners' adware.
</paranoia>