6 ms·

His claim: across various traits, men and women have distributions that differ in mean but still largely overlap. This is a possible non-bias cause of the gende

His claim: across various traits, men and women have distributions that differ in mean but still largely overlap. This is a possible non-bias cause of the gender gap in tech. Specifically to ward off the concern you're about to raise, he adds a tedious note and graphic: http://diversitymemo.com/#possible-causes-gender-gap http://diversitymemo.com/#possible-causes-gender-gap You read all of this, and promptly summarize: "women are, in the the large, excepting some outliers, biologically disfavored to become programmers." What's happening here?

tptacek 9y ago
This isn't responsive to what I wrote.
- brandonhsiao 9y ago
  What? Sure it is. You say he said "Women are, in the the large, excepting some outliers, biologically disfavored to become programmers" when he didn't.
  - elefanten 9y ago
    Aren't those effectively the same claim? If we take his non-bias cause to be true, that directly means that women have biological disadvantages to becoming programmers. He is in fact saying that's why the gap exists.
    - blusterXY 9y ago
      No. He says that genetically-influenced behavioral traits which are observable in aggregate across genders may lead women/men to respond differently to different work environments. Then he suggests a few ways Google can promote diversity without discriminating against anyone. His facts aren't even controversial to anyone who knows the science at this point. Google has basically shown themselves as anti-science and evil.
      - tptacek 9y ago
        I don't understand why people think that reframing things as disputes about "genetically-influenced behavioral traits which are observable in aggregate across genders" is such a powerful argument. We all know what that means. We get it: he's not biased against any one individual woman, just against the statistical distribution women come from. Replacing the word "misogyny" with 10 Latinate jargon words isn't an argument; it's just bad writing.
        blusterXY 9y ago
        Most people don't consider the mere statistical description of aggregate group behavior "misogynous". Nor should they. You use powerful emotional rhetoric when you embrace this rhetoric, but the one condemning the women/men involved for their different aggregate preferences seems to be you. More philosophically, there are compelling reasons to favor "equality of opportunity" and "freedom of choice" over any "equality of outcome" that can only be achieved through coercion and discrimination. At the very least, if someone doesn't want to do something of their own accord, it is hardly doing them a favor to coerce them into changing. And from where comes our moral standing to judge the preferences of other people anyway?
      - ubernostrum 9y ago
        I should start a kickstarter to buy every one of you people a copy of The Mismeasure of Man and force you all to read it.
        blusterXY 9y ago
        > "you people" Do you mean black people or women? Or is this an attack on educated people regardless of their gender or racial orientation or identification? The Mismeasure of Man is a 1981 book by Harvard paleontologist Stephen Jay Gould. The Big-Five model which underpins most of the relevant research in psychology) dates to the mid-1990s. If you want to attack it for being unscientific, you will have to learn about it. Good luck.
        9y ago
    - Digory 9y ago
      No? He's addressing, in part, preferences, that are distributed differently. If some women, on average, want to do things differently than men, on average, that will result in some apparent gap in the averages -- not because of advantage, disadvantage or bias, but because you're looking at a difference in average preference.
    - ameister14 9y ago
      He's saying that's why the gap exists, but he's saying it on an aggregate level, not an individual one. He's not saying all women are disadvantaged, he's saying the proportion of women who are unqualified is greater than the proportion of men who are.
    - rev_null 9y ago
      Except, to establish that it is somehow a biological difference, he would have to rule out cultural bias. Otherwise he's falling back on pseudoscience to justify his position. And, the facts have repeatedly shown that cultural bias is keeping women and minorities out of tech. As an example, I could present a graph showing that most Italians are Catholic and argue that it shows a genetic predisposition. But, such an argument would clearly be absurd.
  - tptacek 9y ago
    My point has in fact very little to do with how you choose to frame the point at issue; rather, it takes issue with the supposed principal underlying it, which is that there should be no "things you cannot say". Clearly there are plenty of things we all (to a first approximation) agree that you cannot in earnest say at your workplace.
    - brandonhsiao 9y ago
      Oops, I missed your point. I still think your summary is not what people are fighting as the threshold issue, but, I agree with your last sentence.
    - mc32 9y ago
      It depends. On the one hand you don't want to sow discord at the workplace, on the other hand you don't want to dismiss critical ideas because the majority don't agree with something. There are other ways to combat illogical ideas. As an atheist I would not fire people for being believers (even if it were lawful to fire on that basis) --just because "I know better". If one of them tried proselytizing, by law I can't do anything about it, but even if I could, there would be better ways to combat irrational appeals and sowing of fear. That said, what if the engineer has "only" presented the ideas as a private person at a "men's club"? If that's not okay, then this infringes upon people's speech outside of work as a citizen --that's chilling. There are Walmarts out there. I believe in making tech an attractive career path for women --as a society we underutilize them and don't extract the value we could --as the soviets did from their women. But at the same time, I am sensitive to kerbing speech.
      - tptacek 9y ago
        This doofus didn't write a memo about keeping people of faith out of Google, and he didn't write it at some secret "men's club". Instead, he had the poor judgement to publish in a workplace forum an argument that the women at Google tended to be there due to reduced standards for hiring women, because biologically women tended to be less suitable for the kind of work they were being hired to do. Timothy Lister had a term for this kind of thing in _Peopleware_: he called it "teamicidal". If you do something teamicidal at your dev job, you should expect repercussions. You are paid to contribute to a team, not to fuck it up in an effort to remake it in your image. (I think Google took the easy way out here, by the way, and that there were more productive ways they could have responded to this, rather than let it fester for days until they were cornered into a dramatic response. But: I don't work at Google, and my epistemological certainty about this is somewhat low.)
UncleMeat 9y ago
He argues that Google should end diversity programs. The document doesn't just poorly summarize social psych and sociology research, but uses it to justify a proposed change in company practice and culture that would perpetuate the status quo. In addition, he showed an incredible lack of judgement in publishing the document. Anybody could tell you that a lot of people would be hurt by this document, regardless of its scientific merit. As such, it makes sense to approach this sort of thing very carefully, as it is possible to be a total asshole and ruin team cohesion while still being "right". Regardless of the validity of these arguments, it should be immediately clear that the document would produce no substantive change from a policy perspective and cause other employees, rightly or wrongly, to feel attacked.
- golemotron 9y ago
  > Regardless of the validity of these arguments, it should be immediately clear that the document would produce no substantive change from a policy perspective That's the problem right there.
- dingo_bat 9y ago
  > Anybody could tell you that a lot of people would be hurt by this document, regardless of its scientific merit. This is a ridiculous statement. Scientific merit should be the only consideration when you are saying something you believe is important. If you're engaged in small-talk, go ahead and throw scientific accuracy out the window, but not in anything substantial.
  - tptacek 9y ago
    There is a very long list of odious things you can claim "science" justifies; in fact, it's kind of hard to think of something you can't six-degrees-of-science your way to defending. This isn't the rhetorical kill-shot you think it is.
    - dingo_bat 9y ago
      > There is a very long list of odious things you can claim "science" justifies; Does the length of such a list invalidate the items in the list? I don't see your point. > in fact, it's kind of hard to think of something you can't six-degrees-of-science your way to defending. Just because something is not directly implied by something else, doesn't mean there is no correlation. If you are working on a mathematical proof, would you give up if you cannot reach the end in 6 (or any arbitrary small number) of steps? Would you declare the assertion wrong? If yes, you're just a bad scientist.

tptacek 9y ago

This isn't responsive to what I wrote.

brandonhsiao 9y ago
What? Sure it is. You say he said "Women are, in the the large, excepting some outliers, biologically disfavored to become programmers" when he didn't.
- elefanten 9y ago
  Aren't those effectively the same claim? If we take his non-bias cause to be true, that directly means that women have biological disadvantages to becoming programmers. He is in fact saying that's why the gap exists.
  - blusterXY 9y ago
    No. He says that genetically-influenced behavioral traits which are observable in aggregate across genders may lead women/men to respond differently to different work environments. Then he suggests a few ways Google can promote diversity without discriminating against anyone. His facts aren't even controversial to anyone who knows the science at this point. Google has basically shown themselves as anti-science and evil.
    - tptacek 9y ago
      I don't understand why people think that reframing things as disputes about "genetically-influenced behavioral traits which are observable in aggregate across genders" is such a powerful argument. We all know what that means. We get it: he's not biased against any one individual woman, just against the statistical distribution women come from. Replacing the word "misogyny" with 10 Latinate jargon words isn't an argument; it's just bad writing.
      - blusterXY 9y ago
        Most people don't consider the mere statistical description of aggregate group behavior "misogynous". Nor should they. You use powerful emotional rhetoric when you embrace this rhetoric, but the one condemning the women/men involved for their different aggregate preferences seems to be you. More philosophically, there are compelling reasons to favor "equality of opportunity" and "freedom of choice" over any "equality of outcome" that can only be achieved through coercion and discrimination. At the very least, if someone doesn't want to do something of their own accord, it is hardly doing them a favor to coerce them into changing. And from where comes our moral standing to judge the preferences of other people anyway?
    - ubernostrum 9y ago
      I should start a kickstarter to buy every one of you people a copy of The Mismeasure of Man and force you all to read it.
      - blusterXY 9y ago
        > "you people" Do you mean black people or women? Or is this an attack on educated people regardless of their gender or racial orientation or identification? The Mismeasure of Man is a 1981 book by Harvard paleontologist Stephen Jay Gould. The Big-Five model which underpins most of the relevant research in psychology) dates to the mid-1990s. If you want to attack it for being unscientific, you will have to learn about it. Good luck.
        9y ago
  - Digory 9y ago
    No? He's addressing, in part, preferences, that are distributed differently. If some women, on average, want to do things differently than men, on average, that will result in some apparent gap in the averages -- not because of advantage, disadvantage or bias, but because you're looking at a difference in average preference.
  - ameister14 9y ago
    He's saying that's why the gap exists, but he's saying it on an aggregate level, not an individual one. He's not saying all women are disadvantaged, he's saying the proportion of women who are unqualified is greater than the proportion of men who are.
  - rev_null 9y ago
    Except, to establish that it is somehow a biological difference, he would have to rule out cultural bias. Otherwise he's falling back on pseudoscience to justify his position. And, the facts have repeatedly shown that cultural bias is keeping women and minorities out of tech. As an example, I could present a graph showing that most Italians are Catholic and argue that it shows a genetic predisposition. But, such an argument would clearly be absurd.
- tptacek 9y ago
  My point has in fact very little to do with how you choose to frame the point at issue; rather, it takes issue with the supposed principal underlying it, which is that there should be no "things you cannot say". Clearly there are plenty of things we all (to a first approximation) agree that you cannot in earnest say at your workplace.
  - brandonhsiao 9y ago
    Oops, I missed your point. I still think your summary is not what people are fighting as the threshold issue, but, I agree with your last sentence.
  - mc32 9y ago
    It depends. On the one hand you don't want to sow discord at the workplace, on the other hand you don't want to dismiss critical ideas because the majority don't agree with something. There are other ways to combat illogical ideas. As an atheist I would not fire people for being believers (even if it were lawful to fire on that basis) --just because "I know better". If one of them tried proselytizing, by law I can't do anything about it, but even if I could, there would be better ways to combat irrational appeals and sowing of fear. That said, what if the engineer has "only" presented the ideas as a private person at a "men's club"? If that's not okay, then this infringes upon people's speech outside of work as a citizen --that's chilling. There are Walmarts out there. I believe in making tech an attractive career path for women --as a society we underutilize them and don't extract the value we could --as the soviets did from their women. But at the same time, I am sensitive to kerbing speech.
    - tptacek 9y ago
      This doofus didn't write a memo about keeping people of faith out of Google, and he didn't write it at some secret "men's club". Instead, he had the poor judgement to publish in a workplace forum an argument that the women at Google tended to be there due to reduced standards for hiring women, because biologically women tended to be less suitable for the kind of work they were being hired to do. Timothy Lister had a term for this kind of thing in _Peopleware_: he called it "teamicidal". If you do something teamicidal at your dev job, you should expect repercussions. You are paid to contribute to a team, not to fuck it up in an effort to remake it in your image. (I think Google took the easy way out here, by the way, and that there were more productive ways they could have responded to this, rather than let it fester for days until they were cornered into a dramatic response. But: I don't work at Google, and my epistemological certainty about this is somewhat low.)

UncleMeat 9y ago

He argues that Google should end diversity programs. The document doesn't just poorly summarize social psych and sociology research, but uses it to justify a proposed change in company practice and culture that would perpetuate the status quo. In addition, he showed an incredible lack of judgement in publishing the document. Anybody could tell you that a lot of people would be hurt by this document, regardless of its scientific merit. As such, it makes sense to approach this sort of thing very carefully, as it is possible to be a total asshole and ruin team cohesion while still being "right". Regardless of the validity of these arguments, it should be immediately clear that the document would produce no substantive change from a policy perspective and cause other employees, rightly or wrongly, to feel attacked.

golemotron 9y ago
> Regardless of the validity of these arguments, it should be immediately clear that the document would produce no substantive change from a policy perspective That's the problem right there.
dingo_bat 9y ago
> Anybody could tell you that a lot of people would be hurt by this document, regardless of its scientific merit. This is a ridiculous statement. Scientific merit should be the only consideration when you are saying something you believe is important. If you're engaged in small-talk, go ahead and throw scientific accuracy out the window, but not in anything substantial.
- tptacek 9y ago
  There is a very long list of odious things you can claim "science" justifies; in fact, it's kind of hard to think of something you can't six-degrees-of-science your way to defending. This isn't the rhetorical kill-shot you think it is.
  - dingo_bat 9y ago
    > There is a very long list of odious things you can claim "science" justifies; Does the length of such a list invalidate the items in the list? I don't see your point. > in fact, it's kind of hard to think of something you can't six-degrees-of-science your way to defending. Just because something is not directly implied by something else, doesn't mean there is no correlation. If you are working on a mathematical proof, would you give up if you cannot reach the end in 6 (or any arbitrary small number) of steps? Would you declare the assertion wrong? If yes, you're just a bad scientist.