But it's not clear how you distinguish between what's an appropriate correction to historical discrimination and what's overcompensation. If the bias were in favor of an over-represented group, it would be more clear -- but that's not what we're seeing. Women are grossly underrepresented in STEM faculty, and the only way to correct that is to hire a greater share of them and to use their gender as a positive discriminating factor in hiring.
I don't believe either of us are in a position to judge whether that should best manifest as a 2:1 ratio during resume reviews, a 1.1:1 ratio in finalized hires, or a 5:1 ratio in outreach, or some other arbitrary number during some other arbitrary phase.