Once again, conversation seems to be ramping up on attention. Ad networks are really interesting when you talk about attention. Due to the reach and the volume of media that networks digest, they gather more data about the value of inventory than anyone. Google, Advertising.com and Yahoo probably know as much about the value of a page or a click as anyone on the Internet. Of course, this is from an advertising perspective and there are many other perspectives for analyzing data, but there are probably no other sources of such broad data sets out there.
Of course, attention value is also relative. Someone querying “Plasma TV” has a different attention value to Circuit City than it does to Cogmap. Sites doing book reviews are probably more interested in the long tail than someone selling books and seeking to drive volume through the big head.
Finally, to assume that every action in a clickstream is equal would be a mistake. One of the great things about networks is that they can differentiate pretty effectively between the value of a good click and a bad click. Integrating data about the network values with clickstream data would be an interesting way to gauge attention value. When you look at things like APML the problem seems to be creating apples to apples comparisons between various metrics of attention value. A broad value engine like Google would let you easily do that.