“Traditional” vs. “Consumer-supported” media: What does it mean?

This Associated Press story says: “U.S. consumers are increasingly shifting their attention away from traditional, advertising-supported media in favor of entertainment such as the Internet, video games and cable TV, which consumers pay for.” It cites a study from Veronis Suhler Stevenson. That private equity firm’s site says:

Consumers are … migrating away from advertising-supported media, such as broadcast TV and newspapers, to consumer-supported platforms, such as cable TV and videogames. Time spent with consumer-supported media grew at a compound annual growth rate of 19.8 percent from 2001 to 2006, while time spent with ad-supported media declined 6.3 percent in the period.

Here’s what I don’t get. Why is cable, for example, not considered an ad-supported medium? Yes, you pay for cable, but you also pay for the newspaper. Both have a lot of ads, and business models that rely on having a lot of ads. Maybe you could make an argument here by breaking out viewership of premium channels and ad-free video-on-demand services. But I don’t think the typical TV viewer (and a majority of American households have cable TV) really makes the implied distinction between, say, Bravo and NBC. When I watch either one, they both look pretty ad-supported to me. And both certainly feel significantly less “consumer supported,” if that is supposed to mean actively spending money, than buying a newspaper does.

What about the AP including the Internet in the not-ad-supported column? Again, yeah, you pay for Internet access, but from everything I’ve read (about what people think will happen at the NYT and WSJ sites, for instance) the movement right now is away from paid-for content and toward content that’s free to the reader — and supported by advertising.

And why aren’t movies and music mentioned at all? Aren’t those relevant to the theory that our entertainment choices are moving away from ad-supported media to things we pay for directly? (But which might of course still include paid product placements — just as many videogames do.) Maybe that stuff is in the full VSS study.

I don’t doubt that media and entertainment consumption patterns are changing, but every report or study I see on the matter seems to have some the data cut some weird way that has nothing to do with how people (as opposed to media companies, I guess) actually consume media and entertainment.

Am I missing something?