If you gave away DRM-free music on an ad-supported site, and found a way to make people want to click those ads, you'd have a good model - providing you found a good way to provide DRM-free music that people wanted. For Facebook, the problem is that they took a long time getting an ad model to work. Their new one is a pretty effective one that takes advantage of the best parts of Facebook. I think they've rounded a difficult corner.
revenue!=profit
Are you aware of the revenue for other social networks? MySpace is going to clear $1 billion in revenue this year, a year faster than Google did. Facebook isn't doing the same numbers but they could, they limit the advertising in order to increase user experience and pursue non-traditional advertising to avoid having their site end up as a friendster. These are strategic decisions, and if it takes them 10 years to fulfill it then, great. That's on them. Maybe that sort of investment secures them 25 or 50 years of high margin profits.
There is no textbook time-line for companies to follow.
I kind of understand the argument that Facebook wasn't running a very good system for revenue: until they released their targeted ad system, they were fairly hit-and-miss and their ads never seemed interesting. Now, they've done what Google did: they made their ads relevant and very much part of their bigger system. I've clicked a few ads, now, because I'm seeing things that I want to click. That's a big breakthrough for Facebook.
Your music site could be monetized in hundreds of various ways. You don't know which traditional ones would work if you don't try. And if you're big enough -- here's the magical part -- you can even get to try non-traditional routes that might just click. That's the whole point of drafting up business models and financial projections and replacing it when need be.
But if you can get 110 million users in 2 years, can cover overhead for 200 top notch engineers and a legion of sales/bizdev team as well as world class infrastructure, hey, I'm not going to be calling you out for not having figured out how to make money on it.
If I create a website tomorrow where I give DRM-free music away legally on an ad-free site I will have 110 million users in 2 years. I will still lose millions a year and my business model will still read [fill in the blank]. Having a lot of users is not an excuse not to have a business model.