First off, "smaller" is comparing May 2014 to Nov 2013, a short-term effect. Second: > The only content type that experienced significant shrinkage was “other

First off, "smaller" is comparing May 2014 to Nov 2013, a short-term effect. Second: > The only content type that experienced significant shrinkage was “other”. > So ruling out third-party content leaves us speculating that either the shrinkage is due to a decrease in use of video (quite possible) or an undocumented change in the testing process (somewhat possible). http://en.wikipedia.org/wiki/Betteridge's_law_of_headlines http://en.wikipedia.org/wiki/Betteridge's_law_of_headlines There is some interesting info in the article, but the headline isn't it.