In some sense it is irrelevant whether or not revenue is independent of price...if you're extrapolating from that dataset, then you really need to give yourself a few good whacks on the head with a stats text. I mean, N=2? Give me a break.
Well, you can't just go changing the price of your app every week just to see what happens. I expect they wrote the blog post to see if anyone else had observed something similar.