That's not right at all. I've had that article pointed out to me before. The author is clueless.
iPhone sales are highly seasonal. A new iPhone gets released towards the end of the year, there's a big spike in sales, which tails off throughout the following year until the next iPhone is released, when the cycle starts again.
What articles like that do is point to the spike at the end of the year, when new iPhones are out and holiday sales are boosting the numbers as well, then point to the following three quarters which are naturally lower (including immediately before a new iPhone is released), and say that sales are "tapering off".
It makes no sense to do that. You aren't comparing like to like. You're comparing the most profitable time of year to the least profitable time of year. Of course sales are going to be lower if you look at it that way - it's true of any product that's seasonal. Would you assume that a suncream company is failing because you looked at their sales in winter and realised that they were selling a fraction of what they do in the summer?
The only sensible way to evaluate sales for seasonal products is to compare year-over-year sales. You compare this year's busy period to last year's busy period. You compare this year's quiet period to last year's quiet period. When you look at the iPhone sales like that - i.e. in the only way that makes sense - sales have never fallen. They have grown every year.