This is an interesting idea, and it builds nicely on your point that you'd rather work for free for 2 hours for the love of helping a friend than work for $5. So if the professional rates drop that the dedicated creators just walk away and make content they love on their own time, then Demand Media and the like face a problem which is that for certain types of content amateurs will produce better results that the (much diminished) professionals.

It would be fun to see a Wiki-like response to Demand Media. One that runs similar algorithms that predict questions people will ask and then asks volunteers to produce better quality results.

Have you seen Jay Rosen's list of sources of news subsidy? There are a lot of options listed there and I think it dovetails nicely with your question about whether third party subsidies of any kind are worthwhile.

The interesting question when you say "a few very rare exceptions in unique niches" is how rare? How unique? Because at the scale of the Internet, very rare is still very large.

I'm reminded of Kevin Kelly's 1000 True Fans argument. We define a true fan as someone willing to pay you one days wage a year for your work. if you have a thousand of them, you end up with 3x their average income as your income. That's a very livable income for a lot of content creators.

I come from a background of videogames, so we have been in a place where we've been competing with free and cross-subsidies since the beginning. And eeking out a living is very hard. But it is doable for some people, and more than one might think.

I think that there are more possibilities.

One is that the price of commodified information is going to crash, but non commodity information will grow in value. So this is stuff like the WSJ and Economist being able to charge for financial information while other papers can't have a paywall. Or Gladwell being asked to give major talks and being paid handsomely.

Another is that it will turn out that there is some minimum threshold of quality that we are willing to accept. If it's true that Google is getting polluted with increasingly low quality results, we should expect a corresponding drop in use around certain topics. Anecdotally, I am already seeing that in my group of friends. We've stopped using Google to do much research about buying because the results for WHATEVER + Review are so predictably bad. Google should respond to this drop by changing how they filter search results, or some other player should rise. I, for one, would love to see the ability to blacklist sites that are routinely non-useful for me from my search results.

A third is that the bad cheap information become so clogging that consumer retreat to high reputation providers. Reviews is one place where I don't use search anymore, I use trusted review sites.

A fourth is that content becomes a kind of loss-leader for other goods and services.