Micropayments
Here's the key quote from McCloud's response to Mr. Shirky:
Suggesting we need fewer choices for our own good betrays a dim view of the user's intelligence
Shirky isn't taking a dim view of the user's intelligence. He's claiming - correctly, in my opinion - that, when faced with two ways of doing something, the user will always choose the easiest option. He's not calling users stupid, he's saying that users prefer usable systems.
This is why people will spend 99 cents for a song on iTunes: Not because the song itself is worth 99 cents, but because it's worth 99 cents to not have to find the songs some other way, to not have to worry about formats, about badly tagged or ripped MP3s and so forth. The content is important, of course, but Apple's iTunes business model only works because the user experience is so breezily convenient.
I pay $9.99/month to eMusic for the same reason: The site has music reviews from music journalists I trust, user comments to help me make decisions about what to download, high-quality rips at good bit rates, etc. I have the technical expertise to skulk around on file sharing services and BitTorrent, but because I get everything I need from eMusic in less time than I would spend otherwise, I generally don't.
So here's a theory. Any delivery service (whether online or off) needs to succeed in doing three things in order to make money. First: clear pricing; second: ease of use; third: ability to prove that the content is worth buying. Both iTunes and eMusic meet these criteria, I think. The iTunes Music Store charges 99 cents per track for all tracks and makes it perfectly evident when you're buying music and when you're only browsing. eMusic has a flat subscription fee. Both services go to pains to show that they provide a long-term time benefit, which pays off the user's initial investment in signing up for the service, and continuing investment in paying for the service.
Free distribution meets the first criterion, and the last, but often fails on the second. Ease of use stems from design (user interface) and infrastructure (no one wants to download large files at slow speed), both of which are far easier to do if you have some kind of budget. Then again, having a budget for infrastructure doesn't necessarily mean you'll have content worth buying.