A colleague (with a young child) brought this topic to my attention. I found the logic behind the idea quite interesting. By the way, the colleague teaches statistics and is female which may in some way explain why she was sensitive to this issue and this approach.
The idea suggests that parents present their young children an interesting math problem at bed time. The web site offers suggestions. Of course, the comparison is to the long standing recommendation from educators that parents read to and with their children from an early age.
My colleague made the observation regarding a difference between reading and math when it comes to our assumptions. We assume “bright” people can read. We do not necessarily assume “bright” people can do math. We assume we work at learning to read (and should start early). We may assume math is an aptitude that you either have or not. I admit I never worked math problems with my kids (at least until later when they had home work) and once they began calculus I could not remember enough to be helpful.