Many efforts at building data infrastructures for the “average researcher” have been funded, designed and in some cases even built. Most of them have limited success. Part of the problem has always been building systems that solve problems that the “average researcher” doesn’t know that they have.