But the think tanks don't get paid for their ideas. And frankly think tanks are a bad example, since they really generate original ideas per se. But usually offer recommendations that are within the realm of what is commonly known in the field, but what's not clear is what is the best solution, e.g., what's the best way to do spectrum auctions.

So in these examples there's usually a handful of ways to handle it. Think tanks will often build models, go through the literature, and recommend a way of doing it. The value isn't in just the recommendation. You can find five other think tanks that will point to five other ways of doing it. The value is in their research.

To bring it back to HN. Here's an idea I have. Wouldn't it be cool to be able to teleport from point A to point B? How much do you want to pay me for the idea? Probably not much. But what if I had done a whole bunch of research in physics that gave a foundation for how it could be done, and built a device that could teleport a particle from A to a random location (so not location B). All of sudden the "idea" looks like it might be worth something. But you're really not buying the idea. You're buying all of the work that comes after the idea. It's not a bad idea, but it's not worth much without a whole lot of work.