Still, RSS readers use plain old HTTP requests (Web Traffic). Transition of users from browser to reader apps shouldn't matter if we count traffic by type.
Right, that was my point (which was written from my phone, so it was terse). RSS is just XML, but to my knowledge, no one's _delivering_ RSS over anything but HTTP in a generic way. Indeed, it's a syndication system _for_ the Web. I will admit, I didn't read the entire article (again, on my phone), but as another commenter pointed out, most of the examples were either pretty-printed data over HTTP, or were services that have never existed in the sphere of the Web.