I think it’s odd that those lines are parallel through the whole time period. I would think that in the late 90s the lines would be closer together (i.e. fewer discouraged unemployed). This analysis seems to indicate (or assume) that the number of discouraged (i.e. the difference between the U6 and Alternate lines) has remained constant since 1994.
But to counteract it with more phoney numbers is just plain wrong.