Brief thoughts on the “Google memo”

Well there is.

So are you suggesting equal efforts should be expended to equalizing the representation of men in nursing and women in bricklaying, as we do other fields? They are skewed worse than most, so perhaps more effort should be put there?

Are the skews in those fields equally unacceptable or somehow more acceptable? And if they are more acceptable, why? Based upon the public discourse, the commonly accepted answer seems to be they are actually more acceptable, since it's rare to hear about efforts for increasing men in nursing or women in construction work relative to other fields like medicine, law, or STEM.

So the question reduces onto "why are some fields with gender skew more acceptable than others?" and answering that question is basically the key to everything.

> Recycling (and slightly modifying) a previous comment

Please don't. There's too much repetition on HN already without copy-pasting previous comments.

If you wrote just the first three paragraphs, I would upvote you. I agree completely. But then:

> Even if there were jobs that women were less qualified for, or had less interest in, we would want lower the difference, with equal representation if possible.

Who are we to tell people what they should do? If people want or don't want to work in some field, they should have the freedom to do so. They shouldn't be forced to work in some field just so that the society could mark the 50:50 checkbox.

Now, you can say, there is "discrimination", and that's why that's somebody preference. But that shouldn't really matter, as long as you fulfill each individual's honest wishes. If I ask you about something, and there is no reason for you to lie to me, then that's your preference and it should be respected if possible. If it's respected, then there is no discrimination, doesn't matter what the group outcome is. Individuals matter, groups are just constructs.

(In fact, the reason why the whole nature vs. nurture debate gets invoked is that somebody has an agenda. You can see the same thing in reverse with gays. Some people want to believe that being gay is cultural, because if it's cultural, then it can be changed, and according to them, should be.)

So this worries me, by going against individual preferences, we are going to make people unhappy, just to satisfy some "higher" goal. There is always a tension like this, in the free market (which I don't believe in), the pay should resolve it, but why to add to it?

Sadly, there are some women out there who are vocal proponents of for example more women programmers (I actually wish for that too, because programming is just cool, but I can't force anybody) and want positive discrimination, but if you ask them personally, if they considered programming as a career, they will tell you that it doesn't interest them for one reason or another. It's incredible hypocrisy.

That's not the argument at all, because there was nowhere in Europe where the government provided no services[0]. Chesterton's fence isn't an argument in favour of non-intervention or laissez-faire economics, it's an argument in favour of inertia. If anything, it implies that the Soviet bureaucracy should have been slowly and carefully dismantled, which isn't obviously wrong if you read accounts of what happened after the dissolution of the USSR.

[0] At least in theory. Failing to provide a promised service is not the same as deliberately providing no service. It might not even possible for a government worth the name to do literally nothing.

> Your analysis assumes that it's always the best candidate that's hired.

Which is something Google has gone on record as saying it does, see for example here: http://www.businessinsider.com/how-google-hires-exceptional-...

But you are correct in that my analysis makes several assumptions that might not hold up in real life, it was meant merely to show that the statement from the original memo doesn't mean (as many people incorrectly thought) that Damore was saying Google was lowering its standards for diversity hires.

> Personally I find it hard to believe that the companiesdoing affirmative action manage to do it without lowering the bar

Ok but even assuming as you do above that a random sample is hired such that "it's still top 25% of women and men that are hired," then you have lowered the bar for everyone from 10% to 25%.

Yes that still meets their own internal requirements, but is what the author meant when he said 'effectively' lower the bar by decreasing the false negative rate.