Did you even read the paper? It seems from this response you've only read other summaries of it. He states numerous times that these analyses don't mean much for the individual, because there is significant overlap between groups.
I did slog through the whole memo.
The way this plays out in the real world is that population differences across genders are also used against individual women who have excellent quantitative or analytical skills on par with the best men. Every mistake, every decision, that these women make is filtered through this prism that women, generally, aren't as good as men. They aren't allowed to make mistakes as highly qualified individuals, rather their mistakes are treated as evidence of their gender's inferiority.
I work in a field that is dominated by women. The topical area we work on isn't really of interest to men. But it's a very quantitative work. I work with many female statisticians, data scientists, and analysts. Many have PhDs from elite schools. I don't think most of the women I work with have a huge, personal interest in our more people-oriented rather than thing-oriented research, they work here because they get to do analytical work without their intellect being constantly questioned due to their gender. Many would be very happy to work in more integrated spaces on more masculine research, and several have, but it just gets so god damn tiring having to constantly work against the assumption that you're not as smart, not as capable, because women "generally" aren't as analytical as men.
So your argument isn't that the author is incorrect, but rather that these ideas can be misused to disqualify women from positions they're otherwise qualified for? Does that mean we just put on blinders and pretend these other factors don't exist?
I get where you're coming from, and you're probably right about how it might be used, but those same people who would use this to disqualify women will use the "diversity hire" excuse to discredit the women they do hire. The tough thing is, if diversity hiring is forced and brings in underqualified candidates, the dissenters will be right because they candidates will actually be inferior to their majority counterparts.
I think this is similar to the armed forces discussions happening over the past few years. Should women be banned from joining elite combat squads like the Army Rangers? No, not if they're qualified. I do find it very hard to believe that of the Ranger candidates, 50% of the qualified personnel are female, so I don't think it should be a forced 50/50 split. If the army forced it to be a 50/50 split, I think they'd put themselves at a disadvantage compared to other similar forces without that restriction.
I guess it depends on how you think we should approach the problem, and where you think we are right now. 30 years ago it was very common for women to be turned down for a position despite being the best candidate. Now it seems at least in some cases (but not all), women are being chosen for positions in spite of not being the best candidate. I get that there's still a pay gap (although far less than the 30% or whatever is usually cited once you correct for work hours and job type), but the pendulum is close to if not past the center point now.
So do we have the Army Rangers force a 50/50 split until the recruits catch up to that ratio, and hope that eventually they do? Or do we try and give everyone an equal shot at things, and accept that being a minority is a given for certain demographics in certain jobs, and something you have to deal with? If you're a guy who wants to be a nurse, or a girl who wants to be a logger, you're going to have to deal with being in the minority.
I'm not sure how one can interpret the overlapping bell curve chart other than to see that the author believes women are inherently inferior at software development.
There are all sorts of reasons why fewer women than men would decide to pursue a career in software. Biological differences are one possibility, yes, but one that is far from proven. As golden1 pointed out earlier, software development was originally seen as "women's work" by men who thought hardware engineering had a higher level of status. Once Microsoft and other software companies started being profitable in their own right, and software development became seen as more of a high-status job, men rushed in.
I'm not discounting the possibility of a small genetic difference in aptitude, but let's be clear. Any biological difference is vastly eclipsed by the social biases around software engineering as an occupation. So many of my women and racial minority colleagues have multiple stories about times when they were explicitly discouraged from pursuing a career in software because a parent or teacher or good friend didn't think it was a good fit for them as a woman or racial minority. Diversity initiatives are meant to combat that type of discouragement by providing explicit encouragement: girls can code, technology is for black people too, etc. To say that possible, unproven genetic differences are a reason to consider stopping diversity initiatives is missing the point entirely.
How about the idea that men are inferior at verbal communication, or understanding people? Is that about women being inferior too?
Men and women are biologically different. There's plenty of overlap, but there has been a lot of research that shows differences in them across not only human cultures, but primates as well. Testosterone alone is a really powerful hormone that changes all sorts of personality traits, ask any guy who's gone on supplementation for it. To say that you somehow know that the difference is just social and completely discount the research is a pretty far reach.
Programs encouraging women to get in to certain classes or hobbies or whatever are a great thing, give them the opportunity to do it. Programs forcing companies to hire candidates based on their gender or race just seem a bit racist/sexist to me, which I thought was the opposite of what we were trying to achieve.