38 comments

[ 3.2 ms ] story [ 93.6 ms ] thread
FTA:

> We know Silicon Valley has a diversity problem.

That's just making an assumption. Based on undergraduate enrollment rates in computer science and other technical degrees, the probability distribution based on ethnicity and gender seem very plausible.

Also FTA,

> Are lawsuits like these a good thing, or are they going to stall efforts to move towards more diversity in Silicon Valley?

Why are we trying to move towards more diversity? Why is that inherently a good thing?

> Why are we trying to move towards more diversity? Why is that inherently a good thing?

For the same reason why genetic bottlenecks are a bad thing.

> For the same reason why genetic bottlenecks are a bad thing.

That is a false analogy. They're segregating based on race and gender, not class or socioeconomic status.

Does it really matter what color your skin is, if you have a MBA from MIT? How does that improve the cultural bottleneck?

"They're segregating based on race and gender, not class or socioeconomic status"

Socioeconomic factors are the more ideal consideration on paper, but in practice it doesn't matter, since that status tends to be influenced very heavily by racial/ethnic factors. Hopefully this'll change someday and we'll all live in a Star Trek utopia. Today is by absolutely no means that day.

"Does it really matter what color your skin is, if you have a MBA from MIT? How does that improve the cultural bottleneck?"

Wait, so just happening to have a degree from a prestigious college suddenly makes someone immune to the effects of racism? What a wonderful thought.

But this is forced diversity on the demand end, instead of genuine diversity on the supply end.

I agree that diversity is a good thing, but hear me out.

What if women are generally genetically predisposed to disliking computer science? I'm not saying they are, but just take it as a given for a moment. Would attempting to increase the diversity in certain places be a good thing to do then? I think it would probably be harmful.

It's unfortunate that we may never know why there are so few women in CS and similar fields. When I saw the 31% stat for women in Silicon Valley, I was surprised, as that seems much higher than I though the general rates were. Maybe that includes managers and stuff?

Believe me, I'd love to see more diversity, but I think forcing it could be dangerous.

"Would attempting to increase the diversity in certain places be a good thing to do then?"

If I'm genetically predisposed to not want to eat my vegetables, does that mean I shouldn't be encouraged to eat my vegetables? Does that also mean that a group of people who are genetically predisposed to want to eat their vegetables shouldn't encourage me to try gardening and possibly even join their vegetable cooking club? Should they not themselves be encouraged to invite me to said vegetable cooking club simply because I'm genetically predisposed to having a tendency to decline such a proposition?

"Maybe that includes managers and stuff?"

More likely customer support, clerical, sales, and other such supporting roles, but yes. That figure is almost certainly not limited strictly or even loosely to technical positions.

>If I'm genetically predisposed to not want to eat my vegetables, does that mean I shouldn't be encouraged to eat my vegetables?

C. S. Lewis said it better than I possibly ever could:

"Of all tyrannies, a tyranny sincerely exercised for the good of its victims may be the most oppressive. It would be better to live under robber barons than under omnipotent moral busybodies. The robber baron's cruelty may sometimes sleep, his cupidity may at some point be satiated; but those who torment us for our own good will torment us without end for they do so with the approval of their own conscience."

Even as a liberal, I find the "It's for their own good" argument frequently advanced by modern authoritarian liberals to be insufferably arrogant. We're all adults here and we deserve the opportunity to chart our own course in life, whether for good or for ill.

I'm by no means trying to imply that such a thing should be forced. On the contrary: it should be encouraged, sure, but the decision to embrace diversity must be a voluntary one. No tyranny wanted or needed here; the benefits of avoiding the stagnation of a monoculture should be apparent enough.
> If I'm genetically predisposed to not want to eat my vegetables, does that mean I shouldn't be encouraged to eat my vegetables?

You're conflating two arguments. First, in the western diet, we objectively eat too little vegetables. It's been proven that eating more vegetables helps your overall well being.

You're begging the question here by equating eating more vegetables in a Western diet to having more diversity. The former has been proven to be good, the latter has not yet been proven to have any effectiveness.

And as another commenter has brought up, should be we forced to do things that are "good" for us?

I know that at Google, you're forced to interview one underrepresented minority candidate for each position you're trying to fill. (Underrepresented meaning black, hispanic, or female) That sounds less like encouragement and more like coercion.

"And as another commenter has brought up, should be we forced to do things that are "good" for us?"

No, and I never stated or even implied such a thing. The choice to embrace diversity is precisely that: a choice. The choice to eat vegetables is precisely that: a choice.

In both cases, there are purported benefits (I'd argue them in both cases to be much more proven than what you seem to believe, but that's another discussion altogether). There's nothing wrong with encouraging the world to explore those benefits. Said encouragement is not tyranny or coercion.

With that said...

"I know that at Google, you're forced to interview one underrepresented minority candidate for each position you're trying to fill. (Underrepresented meaning black, hispanic, or female) That sounds less like encouragement and more like coercion."

I agree completely. Google and other large companies like it do not genuinely understand the benefits of diversity; they are more interested in checking boxes to make themselves look good. Along with that, they fail to realize that embracing diversity must be a voluntary and wholehearted choice for it to be effective.

> For the same reason why genetic bottlenecks are a bad thing.

What is that "genetic bottleneck" ? it's a pretty racist expression if you ask me , it sounds like it leads to "degeneration" or something like that, it really sounds like nazi rhetoric but reversed, instead of the "purity of the race" that must be preserved, it is now mandatory to diversify "white people's genes" to avoid genetic defects ?

You take a Slovenian and an Irish, they are both "white" , does that constitute a "genetic bottleneck" because of a superficial trait ( color of the skin ) ? I mean only racialists speak about genetics when it comes to race. Just because now a specific racial narrative is pushed by the left doesn't make racialism less obnoxious. And white women despite being women are still white , does it also constitute a "genetic bottleneck" too ? people pushing for all that should take a step back and think about what they say, for a second. "Diversity" AKA affirmative action AKA "positive discrimination" isn't inherently good or bad, from a productivity or a financial perspective for businesses that engage into all these diversity policies. If they see value in it, good for them, now let's see if they perform better on the long run just because they replaced a good white candidate with an average non white one.

I'm not american, but I'm puzzled by all this which mostly looks like feelgood PR, seen from abroad, especially at a time US finds more and more difficult to hide its racial tensions to the rest of the world.

The point wasn't to take genetics literally.

The point was that - much like how a homogeneous genetic population will tend to amplify its defects compared to a heterogeneous one - so will a homogeneous ideological population. The diversity is what brings in new ideas and viewpoints.

The very idea of the American "melting pot" is rooted in this philosophy; immigrants both adapt to existing customs and introduce new ones, thus making the America as a whole (lots of us Americans seem to forget this, including at least one of our presidential candidates, but such is life). This can be and has been applied on smaller scales to various organizations with much success.

This isn't only about race or only about gender or only about sexual orientation or only about political affiliation or only about socioeconomic class; it's about all of these things influencing an individual's background in a way that is different from the group as a whole.

> much like how a homogeneous genetic population will tend to amplify its defects compared to a heterogeneous one - so will a homogeneous ideological population. The diversity is what brings in new ideas and viewpoints.

Ahhhh! But we don't know that. Memetics is not genetics.

It's like assuming that a larger population has a larger number of ideas than a smaller population.

But what if I told you my smaller population sample was from a university and the larger one was from a prison.

I believe Dense networks is what causes a flowering of intellectual diversity. There are heuristics we can use. It is associated with wealthy areas and cities.

I suspect myself that intellectual diversity is orthogonal to the physical phenotype of diversity. However this, like your statement, remains to be shown. There is a lot of contradictory information out there, probably due to confounding factors and gloming unrelated factors together.

In my previous post in a different thread I wrote something related to this subject which is that Facebook (users) has a lot of genetic diversity in its population but it's a sterile wasteland. Facebook and Twitter are tourists on the Internet, but 4chan, reddit and HN are like natives.

"Ahhhh! But we don't know that. Memetics is not genetics."

No, but it's a perfectly good starting point for a hypothesis. Memetics may not be genetics, just like how ducks might not be crows. But, like how ducks and crows are both birds, memetics and genetics are both means of conveying/replicating/mutating information.

"But what if I told you my smaller population sample was from a university and the larger one was from a prison."

What if you told me such a thing? Would you be implying that prisoners come up with fewer ideas than universities? From what do you derive such an implication?

"I suspect myself that intellectual diversity is orthogonal to the physical phenotype of diversity."

On paper, it is.

In practice, intellectual characteristics correlate strongly to educational background, which correlates strongly to socioeconomic class, which (at least here in the United States) correlates strongly to - you guessed it - physical phenotype. Some people break out of those correlations, but they're (last I checked) the exception, not the norm.

This is by no means anything inherent in those phenotypes, of course. You already know that. Rather, it's a systemic issue: one of social norms and preconceived notions. We can pretend that everyone's equal, but it ignores the core of the issue: that the segregation of viewpoints still exists, and thus still requires conscious effort to correct that segregation.

Intellectual diversity, in short, correlates with phenotypical diversity, since viewpoints correlate with phenotypes thanks in no small part to our world's rich xenophobic history.

"In my previous post in a different thread I wrote something related to this subject which is that Facebook (users) has a lot of genetic diversity in its population but it's a sterile wasteland."

I'd argue that has far less to do with the users than it does Facebook itself.

"Facebook and Twitter are tourists on the Internet, but 4chan, reddit and HN are like natives."

I don't blame you for believing that; it's natural to believe that we're somehow "special" because we happen to be chatting on some forum run by some venture capital firm. Never mind that even HN's users, let alone reddit's (and possibly 4chan's, but I don't really venture over there all too often), have no problem with posting links to Twitter posts for "karma", or that Medium (a favorite among HNers, so it seems) required logins using either Facebook or Twitter until only very recently.

The flow of thoughts between the "tourists" and the "natives" is very bidirectional in reality, and Facebook and Twitter (and Instagram and $DEITY-knows-what-else) might as well be natives at this point, too.

Great, I'll tell my Peul friend that he's a genetic bottleneck for wanting to marry a Peul girl
>Why are we trying to move towards more diversity? Why is that inherently a good thing?

Nobody's ever been able to show that it is. It's just ideological canon for people pushing quotas.

Your sources are misleading at best. From the first article:

> We know intuitively that diversity matters.

I know nothing of the sort. And that is the lead sentence, which implies the conclusion is already made.

> Companies in the top quartile for racial and ethnic diversity are 35 percent more likely to have financial returns above their respective national industry medians.

Isn't that somewhat expected? Anybody in any quartile for something has a 50% chance of being above national industry medians. If anything, this would cause people to believe that diverse companies underperform.

From the second article:

> Of 366 public companies analyzed, those in the top quartile for racial and ethnic diversity are 35% more likely to have financial returns above national industry medians.

Seems like a regurgitation of the first article, and she presents absolutely nothing to back up those claims. The article quickly devolves into a list of todo items for your company.

From the third article:

> By correlating diversity in leadership with market outcomes as reported by respondents, we learned that companies with 2-D diversity out-innovate and out-perform others. Employees at these companies are 45% likelier to report that their firm’s market share grew over the previous year.

Isn't there a 50% chance that your firm's market share that your market share will grow over the last year? What other variables are you controlling for? And we all know how accurate self reporting is.

> Companies in the top quartile for racial and ethnic diversity are 35 percent more likely to have financial returns above their respective national industry medians.

I'd read this as "In excess of their expected appearance rate" (i.e. 1.35 * 50% likelihood).

> I'd read this as "In excess of their expected appearance rate"

That's a successful rhetorical technique.

There are lies, damned lies and statistics. What if I told you that 40% of sick days were taken on Monday and Friday?

Your brain instantly fills in "because they want a three day weekend", not "well that was expected".

People who write these articles are experts at letting us fill in the blanks.

These studies never seem to have the troubles that other scientific studies have, repeat-ability, outliers or biases at picking companies. If you pick the best of a wider selection, the result is to be better of course, but even then, some companies should screw up just because - the noise of a complex world, where good employees might not overcome bad management or bad circumstances.
Is your argument that this interpretation is numerically wrong or that you don't like the phrasing? It matches my read numerically.

Still post hoc fallacy.

Edit: I'd->is

You seem to be overlooking the word "more" in "35% more likely". It's not "35% chance".

You brain is very good at filling in absent words to change the logical content of a statement.

No, I'd parse that as being 2/5. The difference is, you didn't say "40% more sick days are taken on Monday and Friday", which would imply (M+F) = 1.4(T+W+T) -> 58% of sick days taken on Monday and Friday, and 42% of sick days taken on the remaining days.
There's no empirical proof that diversity is a good thing. Historically people seemed to do fine without any diversity. E.g. most of modern science and technology was discovered/invented by white men. Lack of diversity didn't become a handicap for them.

Disclaimer: I'm a non-white male.

> Lack of diversity didn't become a handicap for them.

How about the opportunity cost of not using the women and non-white people?

most of modern science and technology was discovered/invented by white men. Lack of diversity didn't become a handicap for them

I believe there are probably good arguments for diversity being overvalued. That is not one of them. That would flunk you out of Logic 101.

I'm not sure why this is down voted. Just because X historically achieved A doesn't imply that achievement(X+Y) <= A, so you can't deduce the absence of a handicap from that.
"Groups of diverse problem solvers can outperform groups of high-ability problem solvers" http://www.pnas.org/content/101/46/16385.full

By the way, the paper is a lot of words and math to explain that homogeneous groups can fall into local minima, whereas diverse groups can find better minima. I don't necessarily claim it disproves what you say, but it is at least a paper on the subject.

>Why are we trying to move towards more diversity? Why is that inherently a good thing?

The elite believe that if they reflect the demographics of the general population, the masses will have even less right to get mad about the separation into an elite and an exploited proletariat.

IMO, the idea that anyone could know the optimal level of diversity where human choice is concerned is full of hubris.

Also, "discrimination" is inseparable from choice. That there are laws regulating these things is destructive and arrogant. It turns everyone into racists/sexists/whatever.

the idea that anyone could know the optimal level of diversity where human choice is concerned is full of hubris.

It's completely subjective (how you define "optimal"), so why would it require hubris to have an opinion?

If it's only an opinion about diversity at the dinner table, then it's not hubris. But if the search for an optimal mix comes from a govt. or quasi-govt. institution then it's no longer an opinion, is it?
Asians have been shown repeatedly to be penalized in college admissions (http://observer.com/2015/06/asian-americans-are-indeed-getti...) and in some cases are given handicaps in SAT/ACT scores. Asians have been shown to be the highest earnings among minority groups in America, and perhaps there's a concerted effort by other groups to redistribute the wealth in a way that favors them over Asians.

http://blogs.voanews.com/all-about-america/2015/05/08/this-u...

Men are certainly the beneficiaries of the least amount of "Learn to code" largesse, STEM hiring initiatives, quotas for board members (especially before going public), etc. There are numerous social, governmental and academic programs targeted towards women so while we can't say men are discriminated against, we can say the aid programs that are available for women are nonexistent for men.

People always wants to better their lives. No one is pushing for diversity for coal mining jobs. It's just that Silicon Valley is this hot place where you can make a lot of money so everyone just wants a piece of that. And that includes the most entitled, spoiled and loud peace of shits who instead of thinking how to provide value are pushing for affirmative action. It's not hackers' and geeks' only world anymore. Leeches are here.
According to the diversity reports from linkedin, Facebook, and Google, Asian-Americans are greatly overrepresented in the tech industry. I don't remember the exact numbers, but it's something like 30-40% vs under 10% in the general population.

I don't really know, it may be because Asian American parents are more likely to encourage their children to go into technical careers. Still, a purely numerical statistic is nothing to allege discrimination over, no matter which groups are involved.

As far as Palantir discriminating against Asian Americans goes, my understanding is that most of their "engineering" positions aren't really that technical--they tend to be customer facing. Old timers in government don't really like Asians, and so it "makes sense" to not have them represent the company. (Descriptive of what's happening here, not prescriptive--even if that hypothesis is true, I'm the type to say screw it, hire on capabilities and work ethic.)

There's also the tendency for Asians to be conveniently bucketed into the "white" bucket, which makes them easier targets and also is a reason this lawsuit won't go viral on social media.

> Old timers in government don't really like Asians

Sorry, but I don't believe it's appropriate to make generalizations like this without any evidence. We don't need more blind divisiveness in this (or any) country.