On the other hand, the current system of scientific journals is, in many ways, a complete joke. The demand for referee reports of submitted articles is out of control, and I don't see Arxiv as a solution, as it has its own cultural biases.
I wonder if the problem here is not one of equilibria. Journal articles have become the an important way to measure productivity and to define success. As a result, researchers who are good at producing journal articles prosper. To change the system you have to overcome the intertia of the decision makers often being those reseachers who were successful at the current system.
Most systems of prestige tend to be hard to overturn and it's unclear as to what would be a good pathway to doing so. I tend to wonder about marginal improvements as these are more likely to be adopted. But it is pointless to fight for scientific expression to be done on blog posts if employers count grants and papers to determine future employment.
So what are the improvements at the margin that are possible?