> If the percentage of female computer science graduates (as a strong proxy for the available candidate population)
I don't see why one would assume that gender distribution of CS degree holders is a strong proxy for the gender distribution of available candidates suitable for software development positions. Its true that many big firms (and even a large number of small firms) have a well-known strong bias for CS degree holders (and for CS degree holders from particular elite institutions), but the statistics of the field as a whole, and of CS degree holders, show that that is a bias inconsistent with the real qualifications for the field which cannot avoid exaggerating the gender gap in the field.
> A while back, I did rough envelope calculations that if one of the major FAANG companies hit their diversity goal, there would be none left for any of the rest of them.
That only makes sense if you assume that none of the meeting the target was done by bringing people into the field that otherwise would not be in it, despite the fact that that is typically a focus of such efforts by large firms.
"statistics of the field as a whole, and of CS degree holders, show that that is a bias inconsistent with the real qualifications for the field which cannot avoid exaggerating the gender gap in the field."
You've made a counter-intuitive point and claimed that it's confirmed by statistics and yet haven't included any...
I don't see why one would assume that gender distribution of CS degree holders is a strong proxy for the gender distribution of available candidates suitable for software development positions. Its true that many big firms (and even a large number of small firms) have a well-known strong bias for CS degree holders (and for CS degree holders from particular elite institutions), but the statistics of the field as a whole, and of CS degree holders, show that that is a bias inconsistent with the real qualifications for the field which cannot avoid exaggerating the gender gap in the field.
> A while back, I did rough envelope calculations that if one of the major FAANG companies hit their diversity goal, there would be none left for any of the rest of them.
That only makes sense if you assume that none of the meeting the target was done by bringing people into the field that otherwise would not be in it, despite the fact that that is typically a focus of such efforts by large firms.