The goal here is to avoid having one big meeting where dissent can erupt.
No, the point is to do it in smaller groups so that leakers are less likely to be present (and more easily identified). The meeting was cancelled because people asking questions in the internal question system were being outed -- with their names and locations -- to the public and they were afraid of doxxing and perhaps worse.
A bunch of smaller meetings can be more controlled, more staged, and more scripted.
Smaller meetings will be harder to control or stage, and less able to be scripted. Employees understand that large meetings have to be scripted to a large extent, just for logistics. But in smaller groups employees will feel safer about pressing hard (and the culture in Google is that pressing execs hard is not only allowed, but often rewarded).
if one of them goes south, it won't get as much attention
Maybe... but I doubt it. If one of them goes south, you can bet there will be a raft of memes on the internal memegen, posts on the internal Google+ that will quickly go viral, emails on the many large internal mailing lists, etc.
The only way that wouldn't happen is if many more employees hopped on the internal fora to shout down the complainers. And that could happen. The execs don't have a snowball's chance in hell of controlling employee outrage, though, regardless of group size.
They can use the dissent from any of the smaller forums to form a reactionary, damage control game plan for future sessions as well.
Perhaps, but that's just a bunch of loaded words that say "they can figure out how to make employees happier" (I loaded it the opposite way).
I don't know where you've worked, but every company only wants a monologue. They will tell you they want a dialogue, and talk about "team-building" and "horizontal management structures" and other happy-crappy bullshit.
Google's meeting setup is very much designed to enable a dialogue, to the extent possible. I explained the system here.
The purpose of a "townhall meeting" is dialog. Google had already made it clear that they want a monologue. Cancelling it was very sensible.
Utter nonsense. Did you even read the reason for the cancellation?
The way Google does these 60K-person meetings, to make it possible to have a dialogue, is with a system called "Dory" (named after the Finding Nemo character). Employees log into the system and enter their questions, other employees vote for and against the questions, so the ones most widely viewed as important rise to the top. Then, during the meeting, the execs answer as many of the top questions as possible in the time allotted, live. If there are a lot of highly-ranked questions that they don't get to, the execs answer them in writing after the fact. In addition, if the questions warrant it, various execs are assigned action items to follow up on, and they report their followups directly to the person who asked the question and on the Dory, as well as to the CEO.
The system is far from perfect. Execs can try to tap dance around the question rather than answering it. As a partial remedy against this, the meeting also takes live questions, and the questioners are allowed to -- and often do! -- press the execs when they're unsatisfied with the answers. That option is only available to the relative handful of people present on-site, though. In addition, questions can end at an impasse, where the execs have said what they're going to say, but the questioner doesn't feel like the question has been answered even after back-and-forth (live or via Dory). But my perception is that the execs really do try to answer the questions in good faith, to the best of their ability, and within the constraints they have.
So, no, it's not a real dialogue, but it's certainly not a monologue, and it's about as close to a real dialogue as it's possible to have in such a large group, IMO.
And, normally, the expectation is that anyone is free to ask anything. They should be reasonably respectful, etc., and it must be about relevant things, but outside of that there are no limits. I've seen people stand up and strongly question Larry and Sundar to their faces, and be rewarded for it.
Given the topic of this particular town hall, however, it's likely that many employees would feel uncertain about whether they truly could speak. Less for fear of negative reactions from management as for fear of negative reactions from their peers, but maybe from management as well. I expect the first thing Sundar would have said in the meeting is to clarify that no one would be punished for what they said, but I also expect that lots of people would have reservations about how much they believed him.
Firstly, any stock awarded to an employee in the UK is immediately given a "Fair Market Value" evaluation -- again that's ***on award***. So, if you get given 1,000 GBP worth of stock tomorrow, and you pay nothing for it, the UK will tax you at whatever your income rate is for that 1000 GBP.
Are you sure? The US does it differently, and in a way that is much fairer and seems more consistent with the article.
In the US, the FMV is assigned at vesting, not at date of grant. This is much better, because at vesting some of the shares can be (and generally are) automatically sold and withheld to cover income taxes, reducing the chance that the employee gets ambushed with a huge tax bill.
Also, the structure you describe seems very weird when lined up with the article, because apparently (per you), the stock's value counts as income to the employee when it is granted, and (per the article) it doesn't count as an expense against the company until it vests, and those events have different values. So employee income comes out of thin air on the grant date (because it's not from the company, which hasn't seen a corresponding expense) and then company expense goes nowhere on the grant date (because it's not paid to the employee). It seems like accountants would have to invent some sort of future liability contract to cover the grant and make the books balance.
It's not impossible that you're right, but I'm skeptical.
So when employees cash out they will have to pay tax. In the UK once they convert assets to fiat they will have to pay 20%
FWIW, if you're right about UK law, this works differently in the US. Stock grants are considered regular income to the employee (and an expense to the company) at the point in time when they vest. At time of vesting, a fair market value (FMV) is assessed (usually the stock price at closing on the day before the vesting date, IIRC), and that amount of money is counted as part of the employee's gross income. If the employee holds onto the stock for another year before selling, and the stock price goes up, then the difference between the sale price and the FMV at vesting is considered a capital gain, and taxed at the lower long-term capital gains tax rate (of course, if the employee sells in less than a year, it's a short-term capital gain, and those are taxed as normal income). If the price goes down, it's a long-term capital loss which can offset other long-term capital gains, or perhaps deferred to offset future long-term capital gains.
So, if the stock goes down, will Amazon have to pay higher taxes? That doesn't make much sense..
No. Amazon's taxes should not change no matter what the stock does, they deduct that expense when they pay the employee just like they deduct any salary or bonus. If it goes up the employee pays more taxes on it when they cash it out.
Not if UK income tax law works like US income tax law (and the article implies it does). Such stock grants are not considered income to the employee or an expense to the company until they vest, which happens well after the grant and on some specific schedule.
Interesting. In the US, that would count as ordinary income in the year the stock was given; the tax effects make it an uncommon practice
False, on both counts.
In the US, such stock grants are called Restricted Stock Units (RSUs), and they're actually quite common in large tech companies, and in some other parts of industry. My employer (Google) grants them, and I get them; in fact they constitute about a third of my income.
The way it works is that the grants vest on some schedule. For example, each year after my performance review I'm given another bloc of RSUs, on a four-year vesting schedule. Each month, 1/48th of each grant vests. At the point in time when the stock vests, not when it is granted, it becomes income to the employee and a cost to the company. At vesting time, some of the shares are sold to cover income taxes and the rest are deposited into a brokerage account for the employee.
(Actually, I have asked to have my shares automatically sold upon vesting and the cash deposited into my bank account, then I move it to my brokerage and invest it in index funds, on the theory that my financial future is already tightly bound with Google's success, so I should diversify. I recommend that anyone who receives stock or option grants consider doing the same.)
But at Google the gender gap in tech roles is 80-20, according to their own self-reporting. There's something more systemic going on than the subtle psychological differences between men and women.
Perhaps. This woman argues that the differences...
(Hint: Password complexity rules are a good way to prevent the dumbest of passwords from being used.)
(Hint: Two-factor authentication is so dramatically more secure that you're far better off implementing it and letting people choose the passwords they want.)
No, they didn't. Russia's forced version of egalitarianism didn't allow people to choose the course they found most fulfilling. It largely assigned them roles based on the system's evaluation of their talents and abilities. It also distorted their thinking about what they should want by bombarding them with ideological messages that deliberately negated old gender stereotypes -- regardless of whether or not those stereotypes were what people actually found fulfilling.
As long as there is equal opportunity for anyone who merits it, I will continue to roll my eyes at people hyperventilating over the differences.
That makes perfect sense, if you're not an employer who has noticed that diverse teams perform better. Achieving something like equal representation isn't just about some abstract notion of moral fairness, it's also about having the most productive and creative possible people. Not because women or racial minorities are more productive or creative, but because teams that include them are.
Actually, it seems that the means/variances are indeed extremely different.
What are you measuring? Your examples are not measuring difference in ability, they're measuring counts of people at the tail of some unknown composite distribution. And when you look at the tails of similar normal curves, you find large ratios in the number of people at the same position on the two curves, even though the means and variances are quite close together.
In addition, I think you also have to consider feedback effects. This woman points out that although she was competent in her field she left IT because she eventually realized that she just wasn't like her all-male cohort of colleagues and she didn't like not having more like-minded women around to talk to. Thus, a biological basis for a difference in ability or interest that leads to a, say 65/35 "natural" split may produce a sufficiently lopsided environment that all but the most hardcore female engineers are driven out by the "free-floating testosterone" (her words). So, kernel developers being only 0.8% women may be less because capable and interested women are really *that* rare than because in the completely uncontrolled world of kernel development (there is no HR!) the "boys' club" effect becomes so powerful that few women can stomach it. Indeed, even for men, LKML is known as a place not for the faint-hearted or thin-skinned.
When you factor that in, you should probably assume that if Google has an 80/20 split, the "natural" ratio is actually somewhat less lopsided than that, which argues that the ability/interest means and variances are small.
Google having over 20% female engineers compared to only 0.8% top kernel contributors being female suggests the bias is way above any plausible diversity boost.
I strongly disagree. When you factor in both the boost provided by the fact that we're talking about the extreme end of the curve, and the potential feedback effects, I think the core bias is actually very small -- and that view is supported by the science. Studies do find differences in various characteristics, but they are not large, generally only on the order of 0.1 to 0.2 standard deviations on the mean, and the advantage is not universally male. Variance measurements are harder to make, but those are also typically not that large (though larger variances are almost exclusively on the male side).
This is actively harmful to women who are genuinely skilled -- they are to be likely to be disregarded by their peers as "affirmative action hires".
Here I fully and completely agree. Affirmative action approaches that significantly lower the bar, especially below where the business needs it to be placed, harm everyone. But in the case of Google, that doesn't happen, and it doesn't need to happen. Even Damore's document admitted that Googles "lowering of the bar" consisted only in actions taken to reduce the false negative rate (competent people rejected). Instead, Google focuses on working harder to find women and minorities, and may occasionally give a diversity candidate a second shot at the interview process (though without telling either the interviewers or the hiring committee who evaluates their feedback) where a white male who had a bad day on his interview day is just out of luck.
Another thing that can be done is to apply affirmative action to retention. Companies can work harder to retain the competent non-majority employees, by paying them more, for example. Further, I think this sort of action is completely and totally justifiable with hard, cold business logic. If there is a significant diversity boost, then your non-majority employees are actually more valuable to the company than their equally-skilled majority peers, because while both sets may be equal as individual contributors, the
Hence the long application process asking a lot of things, not job related
What things? I don't recall being asked anything not job-related when I was hired.
they won't hire you if you are not going to fit in the culture
Yes, culture fit is an element of the interview evaluation, but political attitudes definitely do not come into it. It's more about looking to see if people like solving problems, if they're open to discussing alternative solutions, if they seem like nice people (not abrasive, etc.).
You'll see that any kind of racism hurts the person doing the discrimination as he gets an unoptimal result.
Now add another factor to your simulation: Diverse teams are more productive than homogeneous teams. Exact numbers are tough to find, but just for example, assume that a maximally-diverse team is 20% more effective than a single-population team, and that the effect scales linearly.
What you'll find is that unless your means or variances are *very* different for the populations, the diversity "boost" dominates, and affirmative action hiring policies help the person doing the discrimination.
Also important to remember that he was a computer scientist addressing the "science" behind an issue outside his area of expertise.
He isn't a computer scientist, although he was working as a software engineer. His BS was in Molecular Biology and his Master's and (uncompleted) PhD work was in Systems Biology.
But at Google the gender gap in tech roles is 80-20, according to their own self-reporting. There's something more systemic going on than the subtle psychological differences between men and women.
Perhaps. This woman argues that the differences are self-exaggerating, that fields which fewer women are interested in pursuing tend to be male-dominated, which makes them even less attractive to women, which makes them more male-dominated, in a cycle which leads ultimately to a situation where only the women most devoted to the field stay in it.
Read the article, it's well-written and insightful.
This accords as well with the experience of Scandinavian countries who have bent over backwards to ensure not just absolute equality of opportunity, but that everyone has the opportunity to pursue whatever course of education they like and have the talent for. And what they've seen is that rather than fields which are historically dominated by one gender or another equalizing, the ratio has become even more extreme. In Norway, for example, engineering fields tend not to be 50/50, or even 80/20, but 90/10. It appears that when you free people to pursue their own interests, the gender gap increases.
Fun fact: many countries have been successful at eliminating the gap.
Which gap? The only gap you've referenced is math grades. But what has been noted in the most egalitarian countries in the world is not that math grade gaps increase (they don't), but that the ratio of women vs men in various professions is more extreme than in less egalitarian countries.
It seems that though there are some biologically-driven differences in ability, the larger biologically-driven differences are in interest. So, the more freedom you give people to pursue their interests, the larger the gender gap in certain fields.
I have no interest in taking away your choices. You may not be allowed to exercise them on public roads, where they endanger other people, though.
neither will the majority of people
There you're dead wrong. There will be a sizable minority of middle-aged people who will want to drive themselves, but the new generation will come up seeing transportation as a utility. They'll start getting rides to school, to soccer practice, etc., in self-driving cars, without needing the assistance of their parents, and they'll grow up with no interest in driving themselves. I have a 16 year-old son who is already thinking that way; he's questioning whether or not it makes any sense for him to go to the effort of getting a driver's license, since he figures in a few years it won't be needed any more. The elderly, who have lost the reflexes and coordination for driving, will also appreciate it tremendously and adopt it immediately, as will people going home from bars, etc.
It'll take a generation or so after we have good self-driving systems, but within your lifetime (unless you're already in your 60s), manual driving will be rare, and soon after it will be outlawed on public roads as a threat to public safety. Because computers don't get tired, don't get distracted, don't get angry, can remain perfectly aware at all times of everything going on around them in every direction, and can communicate at light speed with all of the other nearby vehicles, and not with crude signals like "I'm probably going to turn", but with detailed information.
Protest all you like, SHOUT all you like, it's going to happen because it makes way too much sense. People who like driving will do it for fun in areas set aside for it, not for transportation.
Exactly like horses today, actually. They're ridden for pleasure and for very narrow, specialized purposes. Manually-driven automobiles will be the same.
TL;DR the experts they selected are on the fringes, the vast majority of experts do not support their views.
Cite? Seriously. I have seen absolutely *zero* argument from experts arguing against either Damore's memo's scientific content (as opposed to the rest), or the four experts in the Quillette article, and I have been looking.
I've seen many comments like yours claiming that debunking exists, but no one seems to be able to point to it.
That a certain group forms the majority does not mean it cannot be discriminated against. In fact, it's quite common to discriminate against the most common group, since that is thought to increase diversity. Doesn't make it any less wrong, though.
We're not talking about right or wrong, we're talking about legal or illegal. Very different thing.
And it is not, in fact, illegal to discriminate against a majority group, giving preference to a protected minority. Indeed, in some contexts it can be illegal *not* to.
Everyone has to pay their own insurance for their own perfect driving
The maker of a fully-autonomous self-driving system should be liable for any accidents caused by the system. It's not collective guilt, it should be built into the cost of the vehicle, and you should not have to pay for your own insurance to cover the manufacturer's liability.
Slashdot Mirror
The goal here is to avoid having one big meeting where dissent can erupt.
No, the point is to do it in smaller groups so that leakers are less likely to be present (and more easily identified). The meeting was cancelled because people asking questions in the internal question system were being outed -- with their names and locations -- to the public and they were afraid of doxxing and perhaps worse.
A bunch of smaller meetings can be more controlled, more staged, and more scripted.
Smaller meetings will be harder to control or stage, and less able to be scripted. Employees understand that large meetings have to be scripted to a large extent, just for logistics. But in smaller groups employees will feel safer about pressing hard (and the culture in Google is that pressing execs hard is not only allowed, but often rewarded).
if one of them goes south, it won't get as much attention
Maybe... but I doubt it. If one of them goes south, you can bet there will be a raft of memes on the internal memegen, posts on the internal Google+ that will quickly go viral, emails on the many large internal mailing lists, etc.
The only way that wouldn't happen is if many more employees hopped on the internal fora to shout down the complainers. And that could happen. The execs don't have a snowball's chance in hell of controlling employee outrage, though, regardless of group size.
They can use the dissent from any of the smaller forums to form a reactionary, damage control game plan for future sessions as well.
Perhaps, but that's just a bunch of loaded words that say "they can figure out how to make employees happier" (I loaded it the opposite way).
I don't know where you've worked, but every company only wants a monologue. They will tell you they want a dialogue, and talk about "team-building" and "horizontal management structures" and other happy-crappy bullshit.
Google's meeting setup is very much designed to enable a dialogue, to the extent possible. I explained the system here.
The purpose of a "townhall meeting" is dialog. Google had already made it clear that they want a monologue. Cancelling it was very sensible.
Utter nonsense. Did you even read the reason for the cancellation?
The way Google does these 60K-person meetings, to make it possible to have a dialogue, is with a system called "Dory" (named after the Finding Nemo character). Employees log into the system and enter their questions, other employees vote for and against the questions, so the ones most widely viewed as important rise to the top. Then, during the meeting, the execs answer as many of the top questions as possible in the time allotted, live. If there are a lot of highly-ranked questions that they don't get to, the execs answer them in writing after the fact. In addition, if the questions warrant it, various execs are assigned action items to follow up on, and they report their followups directly to the person who asked the question and on the Dory, as well as to the CEO.
The system is far from perfect. Execs can try to tap dance around the question rather than answering it. As a partial remedy against this, the meeting also takes live questions, and the questioners are allowed to -- and often do! -- press the execs when they're unsatisfied with the answers. That option is only available to the relative handful of people present on-site, though. In addition, questions can end at an impasse, where the execs have said what they're going to say, but the questioner doesn't feel like the question has been answered even after back-and-forth (live or via Dory). But my perception is that the execs really do try to answer the questions in good faith, to the best of their ability, and within the constraints they have.
So, no, it's not a real dialogue, but it's certainly not a monologue, and it's about as close to a real dialogue as it's possible to have in such a large group, IMO.
And, normally, the expectation is that anyone is free to ask anything. They should be reasonably respectful, etc., and it must be about relevant things, but outside of that there are no limits. I've seen people stand up and strongly question Larry and Sundar to their faces, and be rewarded for it.
Given the topic of this particular town hall, however, it's likely that many employees would feel uncertain about whether they truly could speak. Less for fear of negative reactions from management as for fear of negative reactions from their peers, but maybe from management as well. I expect the first thing Sundar would have said in the meeting is to clarify that no one would be punished for what they said, but I also expect that lots of people would have reservations about how much they believed him.
Firstly, any stock awarded to an employee in the UK is immediately given a "Fair Market Value" evaluation -- again that's ***on award***. So, if you get given 1,000 GBP worth of stock tomorrow, and you pay nothing for it, the UK will tax you at whatever your income rate is for that 1000 GBP.
Are you sure? The US does it differently, and in a way that is much fairer and seems more consistent with the article.
In the US, the FMV is assigned at vesting, not at date of grant. This is much better, because at vesting some of the shares can be (and generally are) automatically sold and withheld to cover income taxes, reducing the chance that the employee gets ambushed with a huge tax bill.
Also, the structure you describe seems very weird when lined up with the article, because apparently (per you), the stock's value counts as income to the employee when it is granted, and (per the article) it doesn't count as an expense against the company until it vests, and those events have different values. So employee income comes out of thin air on the grant date (because it's not from the company, which hasn't seen a corresponding expense) and then company expense goes nowhere on the grant date (because it's not paid to the employee). It seems like accountants would have to invent some sort of future liability contract to cover the grant and make the books balance.
It's not impossible that you're right, but I'm skeptical.
So when employees cash out they will have to pay tax. In the UK once they convert assets to fiat they will have to pay 20%
FWIW, if you're right about UK law, this works differently in the US. Stock grants are considered regular income to the employee (and an expense to the company) at the point in time when they vest. At time of vesting, a fair market value (FMV) is assessed (usually the stock price at closing on the day before the vesting date, IIRC), and that amount of money is counted as part of the employee's gross income. If the employee holds onto the stock for another year before selling, and the stock price goes up, then the difference between the sale price and the FMV at vesting is considered a capital gain, and taxed at the lower long-term capital gains tax rate (of course, if the employee sells in less than a year, it's a short-term capital gain, and those are taxed as normal income). If the price goes down, it's a long-term capital loss which can offset other long-term capital gains, or perhaps deferred to offset future long-term capital gains.
So, if the stock goes down, will Amazon have to pay higher taxes? That doesn't make much sense..
No. Amazon's taxes should not change no matter what the stock does, they deduct that expense when they pay the employee just like they deduct any salary or bonus. If it goes up the employee pays more taxes on it when they cash it out.
Not if UK income tax law works like US income tax law (and the article implies it does). Such stock grants are not considered income to the employee or an expense to the company until they vest, which happens well after the grant and on some specific schedule.
In the US, these are called Restricted Stock Units. http://www.investopedia.com/te...
Interesting. In the US, that would count as ordinary income in the year the stock was given; the tax effects make it an uncommon practice
False, on both counts.
In the US, such stock grants are called Restricted Stock Units (RSUs), and they're actually quite common in large tech companies, and in some other parts of industry. My employer (Google) grants them, and I get them; in fact they constitute about a third of my income.
The way it works is that the grants vest on some schedule. For example, each year after my performance review I'm given another bloc of RSUs, on a four-year vesting schedule. Each month, 1/48th of each grant vests. At the point in time when the stock vests, not when it is granted, it becomes income to the employee and a cost to the company. At vesting time, some of the shares are sold to cover income taxes and the rest are deposited into a brokerage account for the employee.
(Actually, I have asked to have my shares automatically sold upon vesting and the cash deposited into my bank account, then I move it to my brokerage and invest it in index funds, on the theory that my financial future is already tightly bound with Google's success, so I should diversify. I recommend that anyone who receives stock or option grants consider doing the same.)
https://youtu.be/tiJVJ5QRRUE?t... (the relevant bit is only about 90 seconds, from 30:57 to 32:24, watch it.)
And so...the gender gap for Computer Science majors is...?
... just another problem to be solved for companies like Google who wish to reap the rewards of a diverse employee population.
But at Google the gender gap in tech roles is 80-20, according to their own self-reporting. There's something more systemic going on than the subtle psychological differences between men and women.
Perhaps. This woman argues that the differences...
Hmm. I just noticed that I inadvertently pasted the wrong link there. The correct one is: https://www.bloomberg.com/view...
(Hint: Password complexity rules are a good way to prevent the dumbest of passwords from being used.)
(Hint: Two-factor authentication is so dramatically more secure that you're far better off implementing it and letting people choose the passwords they want.)
Counterpoint: Russia, which found the reverse.
No, they didn't. Russia's forced version of egalitarianism didn't allow people to choose the course they found most fulfilling. It largely assigned them roles based on the system's evaluation of their talents and abilities. It also distorted their thinking about what they should want by bombarding them with ideological messages that deliberately negated old gender stereotypes -- regardless of whether or not those stereotypes were what people actually found fulfilling.
As long as there is equal opportunity for anyone who merits it, I will continue to roll my eyes at people hyperventilating over the differences.
That makes perfect sense, if you're not an employer who has noticed that diverse teams perform better. Achieving something like equal representation isn't just about some abstract notion of moral fairness, it's also about having the most productive and creative possible people. Not because women or racial minorities are more productive or creative, but because teams that include them are.
Actually, it seems that the means/variances are indeed extremely different.
What are you measuring? Your examples are not measuring difference in ability, they're measuring counts of people at the tail of some unknown composite distribution. And when you look at the tails of similar normal curves, you find large ratios in the number of people at the same position on the two curves, even though the means and variances are quite close together.
In addition, I think you also have to consider feedback effects. This woman points out that although she was competent in her field she left IT because she eventually realized that she just wasn't like her all-male cohort of colleagues and she didn't like not having more like-minded women around to talk to. Thus, a biological basis for a difference in ability or interest that leads to a, say 65/35 "natural" split may produce a sufficiently lopsided environment that all but the most hardcore female engineers are driven out by the "free-floating testosterone" (her words). So, kernel developers being only 0.8% women may be less because capable and interested women are really *that* rare than because in the completely uncontrolled world of kernel development (there is no HR!) the "boys' club" effect becomes so powerful that few women can stomach it. Indeed, even for men, LKML is known as a place not for the faint-hearted or thin-skinned.
When you factor that in, you should probably assume that if Google has an 80/20 split, the "natural" ratio is actually somewhat less lopsided than that, which argues that the ability/interest means and variances are small.
Google having over 20% female engineers compared to only 0.8% top kernel contributors being female suggests the bias is way above any plausible diversity boost.
I strongly disagree. When you factor in both the boost provided by the fact that we're talking about the extreme end of the curve, and the potential feedback effects, I think the core bias is actually very small -- and that view is supported by the science. Studies do find differences in various characteristics, but they are not large, generally only on the order of 0.1 to 0.2 standard deviations on the mean, and the advantage is not universally male. Variance measurements are harder to make, but those are also typically not that large (though larger variances are almost exclusively on the male side).
This is actively harmful to women who are genuinely skilled -- they are to be likely to be disregarded by their peers as "affirmative action hires".
Here I fully and completely agree. Affirmative action approaches that significantly lower the bar, especially below where the business needs it to be placed, harm everyone. But in the case of Google, that doesn't happen, and it doesn't need to happen. Even Damore's document admitted that Googles "lowering of the bar" consisted only in actions taken to reduce the false negative rate (competent people rejected). Instead, Google focuses on working harder to find women and minorities, and may occasionally give a diversity candidate a second shot at the interview process (though without telling either the interviewers or the hiring committee who evaluates their feedback) where a white male who had a bad day on his interview day is just out of luck.
Another thing that can be done is to apply affirmative action to retention. Companies can work harder to retain the competent non-majority employees, by paying them more, for example. Further, I think this sort of action is completely and totally justifiable with hard, cold business logic. If there is a significant diversity boost, then your non-majority employees are actually more valuable to the company than their equally-skilled majority peers, because while both sets may be equal as individual contributors, the
Hence the long application process asking a lot of things, not job related
What things? I don't recall being asked anything not job-related when I was hired.
they won't hire you if you are not going to fit in the culture
Yes, culture fit is an element of the interview evaluation, but political attitudes definitely do not come into it. It's more about looking to see if people like solving problems, if they're open to discussing alternative solutions, if they seem like nice people (not abrasive, etc.).
You'll see that any kind of racism hurts the person doing the discrimination as he gets an unoptimal result.
Now add another factor to your simulation: Diverse teams are more productive than homogeneous teams. Exact numbers are tough to find, but just for example, assume that a maximally-diverse team is 20% more effective than a single-population team, and that the effect scales linearly.
What you'll find is that unless your means or variances are *very* different for the populations, the diversity "boost" dominates, and affirmative action hiring policies help the person doing the discrimination.
Also important to remember that he was a computer scientist addressing the "science" behind an issue outside his area of expertise.
He isn't a computer scientist, although he was working as a software engineer. His BS was in Molecular Biology and his Master's and (uncompleted) PhD work was in Systems Biology.
But at Google the gender gap in tech roles is 80-20, according to their own self-reporting. There's something more systemic going on than the subtle psychological differences between men and women.
Perhaps. This woman argues that the differences are self-exaggerating, that fields which fewer women are interested in pursuing tend to be male-dominated, which makes them even less attractive to women, which makes them more male-dominated, in a cycle which leads ultimately to a situation where only the women most devoted to the field stay in it.
Read the article, it's well-written and insightful.
This accords as well with the experience of Scandinavian countries who have bent over backwards to ensure not just absolute equality of opportunity, but that everyone has the opportunity to pursue whatever course of education they like and have the talent for. And what they've seen is that rather than fields which are historically dominated by one gender or another equalizing, the ratio has become even more extreme. In Norway, for example, engineering fields tend not to be 50/50, or even 80/20, but 90/10. It appears that when you free people to pursue their own interests, the gender gap increases.
An interesting exploration of this issue in Norway is presented in https://www.youtube.com/watch?...
Fun fact: many countries have been successful at eliminating the gap.
Which gap? The only gap you've referenced is math grades. But what has been noted in the most egalitarian countries in the world is not that math grade gaps increase (they don't), but that the ratio of women vs men in various professions is more extreme than in less egalitarian countries.
It seems that though there are some biologically-driven differences in ability, the larger biologically-driven differences are in interest. So, the more freedom you give people to pursue their interests, the larger the gender gap in certain fields.
You will NOT take away MY CHOICES.
I have no interest in taking away your choices. You may not be allowed to exercise them on public roads, where they endanger other people, though.
neither will the majority of people
There you're dead wrong. There will be a sizable minority of middle-aged people who will want to drive themselves, but the new generation will come up seeing transportation as a utility. They'll start getting rides to school, to soccer practice, etc., in self-driving cars, without needing the assistance of their parents, and they'll grow up with no interest in driving themselves. I have a 16 year-old son who is already thinking that way; he's questioning whether or not it makes any sense for him to go to the effort of getting a driver's license, since he figures in a few years it won't be needed any more. The elderly, who have lost the reflexes and coordination for driving, will also appreciate it tremendously and adopt it immediately, as will people going home from bars, etc.
It'll take a generation or so after we have good self-driving systems, but within your lifetime (unless you're already in your 60s), manual driving will be rare, and soon after it will be outlawed on public roads as a threat to public safety. Because computers don't get tired, don't get distracted, don't get angry, can remain perfectly aware at all times of everything going on around them in every direction, and can communicate at light speed with all of the other nearby vehicles, and not with crude signals like "I'm probably going to turn", but with detailed information.
Protest all you like, SHOUT all you like, it's going to happen because it makes way too much sense. People who like driving will do it for fun in areas set aside for it, not for transportation.
Exactly like horses today, actually. They're ridden for pleasure and for very narrow, specialized purposes. Manually-driven automobiles will be the same.
Nonsense. Necessity trumps risk, as long as the risk is sufficiently small to be acceptable.
Sorry, lives trump feelings.
That site was debunked last time.
TL;DR the experts they selected are on the fringes, the vast majority of experts do not support their views.
Cite? Seriously. I have seen absolutely *zero* argument from experts arguing against either Damore's memo's scientific content (as opposed to the rest), or the four experts in the Quillette article, and I have been looking.
I've seen many comments like yours claiming that debunking exists, but no one seems to be able to point to it.
That a certain group forms the majority does not mean it cannot be discriminated against. In fact, it's quite common to discriminate against the most common group, since that is thought to increase diversity. Doesn't make it any less wrong, though.
We're not talking about right or wrong, we're talking about legal or illegal. Very different thing.
And it is not, in fact, illegal to discriminate against a majority group, giving preference to a protected minority. Indeed, in some contexts it can be illegal *not* to.
Everyone has to pay their own insurance for their own perfect driving
The maker of a fully-autonomous self-driving system should be liable for any accidents caused by the system. It's not collective guilt, it should be built into the cost of the vehicle, and you should not have to pay for your own insurance to cover the manufacturer's liability.