Default priors

As I'm making my way through old Gelman posts, I found this post which might particularly be interesting to students in my STAT 615 class (Advanced Bayesian methods) where he discusses default prior choices for hierarchical models. His preference is for weakly informative priors, e.g. t, half-t, and Ga(2,1A) for non-varying parameters, variance parameters, and hierarchical scale parameters.