1 - Market cap is (theoretically) the Net Present Value (NPV) of future cash flows. You have to be careful about comparing this NPV (discounting forever) with a 1 year slice. When interest rates are low, the future cash flows have a lot of value.
2 - Facebook's value is likely based on other things too. Advertising is just 1 way to monetize their users. Imagine hijacking all of voice communications as an example.
So to summarize, the market value of facebook assumes 1) It becomes co-dominant web-advertiser with google in the US
AND
2) It finds another way to make some money (either by other advertising markets OR other products/features that people pay for).
The other thing to consider is the competitive/strategic landscape. It's not so much that the market 'assumes' that Facebook will invent new ways of making money, but rather that regardless of who figures it out, facebook is so incredibly well positioned at the moment (ubiquitous social network) as to give it a large competitive advantage, and let it eat other companies' lunch.
The market has examples of other similarly well-positioned companies, like Microsoft in the early 90s, to draw on.
1) It dominates advertising for the long future
AND
(
2a) It's discount rate (US govt borrowing rate plus a risk premium) is very low
OR
2b) It finds another way to make money )
Facebook's main advantage is a treasure trove of real data about real people, who use their real names.