We don't have that data. You are quibbling on sample size but making no argument on effects. Make the case that this is a good year for Microsoft.