Note that the problem the paper is about is that in real life, using precision instruments, the Feynman's sprinkler does rotate the other way under suction.

The counterintuitive asymmetry between blowing and sucking is already well understood.