Serious question: why does anyone still listens to this hack?

The majority of his results are either questionable or completely obvious, and he's clearly not very good at accounting for his own biases. Some of his work has set back the field of design quite a bit - like the nonsense about users not reading on the internet that has transformed itself into an unkillable monster.

p.s.: if you're coming to say "but dude, users really don't read on the internet" then reflect on the irony of doing that in a discussion forum.