Apparently Nielssen is reporting a small drop in the number of people in the UK using Facebook. 8.9m users in December dropped to 8.5m users in January. Guys, this almost couldn't matter less. The stat we're looking at here is the number of people who visited the site once in January compared to the number who did so in December. That's not a remotely sensible way of measuring Facebook's popularity. How many total minutes did UK Internet users spend on the site in January? (answer: many more than in December.) How many minutes were spent on the site per person in January? (again: more than in December.) How many times did each visitor go to the site in January? (and so on, ad infinitum.)
So some of Facebook's casual, drive-by traffic has dried up (presumably because absolutely everyone with a computer has now looked at it at least once and either said "nah" or "go on then"). No big deal there. Actual engagement is up across the board.
Put it this way: if you're not smart enough to have stopped applying irrelevant mass media consumption figures to this stuff about two years ago, at least be smart enough to apply the right irrelevant mass media consumption figures. On those terms, Facebook's engagement numbers grew in January. The site isn't in decline.
Don't get me wrong: Facebook is still evil and doomed. Just...not in that way.