Atenção: As questões de números 16 a 20 baseiam-se no texto abaixo.

    Technology and legal pressure have changed spammers’ terms of trade. They long relied on sending more e-mails from more computers, knowing that some will get through. But it is hard to send 100m e-mails without someone noticing. In 2008 researchers from the University of California at Berkeley and San Diego posed as spammers, infiltrated a botnet and measured its success rate. The investigation confirmed only 28 “sales” on 350m e-mail messages sent, a conversion rate under .00001%. Since then the numbers have got worse.
    But spammers are a creative bunch. KIK of tricking consumers into a purchase, they are stealing their money directly. Links used to direct the gullible to a site selling counterfeits. Now they install “Trojan” software that ransacks hard drives for bank details and the like.
    Spammers also have become more sophisticated about exploiting trust. In few places is it granted more readily than on social-networking sites. Twitter, a forum for short, telegram-like messages, estimates that only 1% of its traffic is spam. But researchers from the University of California at Berkeley and the University of Illinois at Champaign-Urbana show that 8% of links published were shady, with KIIK of them leading to scams and the rest to Trojans. Links in Twitter messages, they found, are over 20 times more likely to get clicked than those in e-mail spam.
    Nor is Facebook as safe as it seems. As an experiment, BitDefender, an online-security firm, set up fake profiles on the social network and asked strangers to enter into a digital friendship. They were able to create as many as 100 new friends a day. Offering a profile picture, particularly of a pretty woman, increased their odds. When the firm’s researchers expanded their requests to strangers who shared even one mutual friend, almost half accepted. Worse, a quarter of BitDefender’s new friends clicked on links posted by the firm, even when the destination was obscured.

(Adapted from http://www.economist.com/node/17519964)