Well, that is what the article claims, they compare their own stats with comScore and Alexa stats. But I think the conclusion is indeed flawed as the Flurry sta

Well, that is what the article claims, they compare their own stats with comScore and Alexa stats. But I think the conclusion is indeed flawed as the Flurry stats include people with smart phones with apps on it that use Flurry, which is just a subset of all people using the Internet. They should include the absolute numbers to make the story complete. But yeah, there is a strong trend we can't deny :)