> the author of the article you link is "a Microsoft Contributor at Ars"; I am not surprised that he would manage to find good numbers for IE
While I agree that different websites can have vastly different distributions of browsers I am disappointed in your unsupported assertion that just because someone writes news about Microsoft that they have lied about the browser distribution numbers. I think Ars has a little more journalistic integrity than that. I know it is fashionable to make these kind of serious accusations in a completely off-hand manner but it provides a confusing and inaccurate picture of the world, making it more difficult to pick out and recognize the real biases, by using rational, logical thought.
I wish more people learned rhetoric so that they were better equipped to spot BS and didn't fall back to using such inaccurate methods.