A lot of clumping happens in fields. Tell me why women are better at symplectic geometry than symplectic topology (or not). Tell me why women are better combinatorists than number theorists -- I want to hear something about counting the tubers they gathered. Tell me why women are so much more suited to advanced heart failure and transplant cardiology than interventional cardiology (https://www.aamc.org/data/448482/b3table.html). I kinda love all the post hoc analysis that goes into explaining these differences, especially when you have to explain then why Portuguese and Italian women are great in algebraic geometry but French women aren't (except for the notable exceptions like Claire Voisin).

Most people aren't pioneers; most people follow the pioneers into situations in which they feel mildly more comfortable than the alternative. Sexism plays a huge role, while not being the only factor. Is it really surprising that if you put up a "girls r dum" sign on a clubhouse many will find somewhere else to go?

Women electing into different fields is pretty much tautological here. The question is why, and whether it's for reasons we want to support and/or perpetuate. The whole point of the memo at issue is that he figures it's built into women biologically because computers aren't cuddly and that it is thus justifiable and right to perpetuate or support the current set of social roles. I don't know that the guy should have been fired, but it's clear he's not going to be a great team player or supervisor for women.

This article [0] presents a good argument about how and why men and women self-select in to different fields.

0: http://slatestarcodex.com/2017/08/07/contra-grant-on-exagger...