I've been working on a post roughly on this topic for a while, and it's been getting long and meandering - so I've split it up. I'm expecting the following to be part one of several.
Scholarly communication as it is now is far from ideal. In what ways could it change to make things better? One way I have been thinking about it recently is in the context of Stewart Brand's "pace layers diagram"; Long Now Foundation just posted an interesting audio recording of a discussion about its origin and uses. The idea is that systems are composed of layers with differing rates of change:
The fast layers at the top are where innovation and experimentation happens, while the slow layers at the bottom bring stability. Changing things too quickly at the bottom isn't safe; forcing the top to slow down is equally harmful. Even our nature as human beings changes slowly over time, but for the most part that layer is given. Let's examine the other layers of the scholarly communication system a little: