Hacker News Reader

Comment by SenAnder | Hacker News Reader

We've come a long way since the Civil Rights Act of 1964. When it was debated, the great liberal Democrat Hubert H. Humphrey himself stood on the Senate floor during debate on the measure and declared: “I will start eating the pages” of the law if anyone can find a clause that calls for quotas, preferences or racial balance in jobs or education - https://www.latimes.com/archives/la-xpm-1995-02-20-mn-34094-...

Of course, there are still no quotas. Just that, if you don't meet the non-existent quotas, that is evidence for the illegal discrimination prosecution against you.

View on HN

> there are still no quotas

There are no explicit quotas. Every time someone is held out as not being diverse enough, there is a quota implied.

It looks like this is just a reporting requirement and no quotas are mandated. I imagine the intent is to indirectly push companies to improve diversity through embarrassment by public numbers.

However, these are private individuals funding private ventures with private money, as far as I know, so why the government should be involved at all in that process other than taxation isn't clear to me.

This is one step in a long march. If you can remember this march, the next step is obvious.

The Civil Rights Act didn't directly ban using IQ tests for hiring/promotions, yet that's how the courts eventually interpreted it [1]. More recently, a test specifically made to evaluate teachers was declared racist and discriminatory [2], mostly based on disparate pass rates, but in part also on requiring some "cultural knowledge". Ironic, since imparting culture is a large part of a teacher's job - outside of STEM, nearly all knowledge is culturally biased (i.e. there is greater emphasis on US history in the US than in Poland).

[1] https://en.wikipedia.org/wiki/Griggs_v._Duke_Power_Co.

[2] https://web.archive.org/web/20220714122944/https://www.wsj.c...

I'm not really sure how you read those and think the shady activity is on the part of the government. You're missing the left foot of the march which is, "people discover an 'unbiased' assessment that gives them the results they want and then standardize on it." And then the right foot is the courts not being stupid and smacking it down for obvious racism.

I guarantee you if IQ wasn't a measure that affluent white privately educated men did well at it would have never been adopted as a measure.

Asians do better on IQ, Asians do better in college. What happened wasn't that we stopped caring about IQ or college, instead those who oppose them started treating Asians as white and saying they don't count.

So it seems like the people who change like the wind here aren't those in favor of merits.

>I guarantee you if IQ wasn't a measure white people did well at it would have never been adopted as a measure.

Even scientists who say the black-white IQ gap is not in part due to genetic factors agree that it does measure intelligence to some extent: that is, people with higher intelligence do better on average on IQ tests, even if each IQ point is not a numeral unit adding to your intelligence 'score.' All this said, why would you assume that people using this tests only are doing so for biased reasons?

it's still not clear what IQ measures, in the grand scheme of things. and i think this is the bulk of its problem. what sort of intelligence? relevant for what? unfortunately, the laziest among us who see these standardize tests are easy way to do their job embraced it wholeheartedly and have deployed it wickedly. at what point should an IQ test be taken? and what's its expiry date? mr. cipolla wrote his seminal basic laws of human stupidity[0] some 70 years after IQ tests were introduced. perhaps not important in the grand scheme of things but one wonders why there's not even as much as a mere allusion.

[0]: http://harmful.cat-v.org/people/basic-laws-of-human-stupidit...

We don't know exactly what it measures, but we are not surprised that myelination[0] and dendrite length[1] increase with IQ.

[0]https://www.nature.com/articles/s41380-017-0001-5

[1]https://www.nature.com/articles/s41467-023-39946-9

I'll invoke Godwin given the obvious connection, the Jews "only" had to wear stars. Similar situations happen again, and again, and again throughout history, and those that forget history...

More direct comparison would be Jews were excluded from Colleges for the same reason Asians are excluded today https://en.wikipedia.org/wiki/Universities_and_antisemitism Fears of overrepresentation. Doing too well on the grading standards

Explicit quotas would probably be preferable to the current state of affairs. The fact that they're implicit means that, as soon as you bring them up, someone can play dumb and claim that they don't exist. It precludes having an honest discussion or even knowledge of how distorting the quotas are.

Implicit quotas are preferable; if it really matters someone can ignore them. People try strategies of "we'll force the situation to be broken and then they'll have to fix it!" every so often in politics. It often doesn't work with terrible consequences. People lack the imagination to look at a bad situation and easily come to a consensus about how to improve it.

It is much better to enforce things by social expectation. Then enforcement measures can be changed piecemeal.

> Implicit quotas are preferable;

Until you make enemies with a public prosecutor or someone politically well-connected. Then all the legal ambiguity is turned against you.

>Of course, there are still no quotas.

There are, at least in US immigration. It's just not something Americans need to deal with, so it goes under the radar.

If I say "your race/gender distribution should at least vaguely resemble that of the general population" is that a quota?

Is it a quota if I say a fair coin should yield heads at least 49% of the time? What about if I demand that a six-sided die have a normal distribution and not be weighted for cheating purposes?

Thank you! We need to be training female nurses and teachers to be truck drivers and male truck drivers to be teachers. Only then will we have a fair society.

I'll assume you're not arguing against letting women be nurses/teachers, and are instead arguing that we need more male nurses and teachers.

In which case, yeah? Gender bias in medicine has proven negative effects both for patients and physicians. If we do everything right to knock down barriers, we should eventually see more male nurses and more male teachers. Why shouldn't we? What's wrong with a man being a nurse?

This all depends on what "should" means, and what kind of carrot/stick is hiding behind that phrase.

Yes. People are not coins or dice. Judge them by the whole person they are, not the bucket others put them in.

So is this an argument against fairness? Are you saying that it's OK for me to discriminate against you based on your race or gender, since those are parts of who you are as a "whole person"? Or do parts of you not count as the whole person? Should I be able to discriminate against you based on any disabilities you have through no fault of your own, since those are parts of your "whole person" and can limit your job performance?

If a large pool of people like "we have a thousand employees" has a demographic balance that is wildly different from gen pop you at least need a good explanation for how it got that way. There are plenty of ways to explain it, but it's unreasonable to just assume "it's fine, I'm sure nothing bad happened" on faith.

‘If a large pool of people like "we have a thousand employees" has a demographic balance that is wildly different from gen pop you at least need a good explanation for how it got that way.‘

This is a control freak / bureaucrat / authoritarian dream. Endless work to keep up with ever moving targets, and since everyone will be out of compliance at various times it means it’s ripe for selective enforcement.

Also, do you believe that different cultures/groups of people may have different preferences? Is it crazy to think that may affect their career choices and preferences? Should the state then mandate them to enter a profession at a higher rate?

This is something you'd at least do a little research to demonstrate instead of just expecting people to accept, on faith, that "our company is 5% female because of ladies' preferences", or things of that nature.

I said "wildly different" on purpose, we're not talking about "your workforce is 45% female instead of 47% female" here. It really is big differences - like if you pull up a MS diversity report they're around 30% female when the US population is ~50% female. This is a big discrepancy. You can come up with lots of plausible explanations but if you want to be believed, shouldn't you have a little data? Or should we just shrug off the potential of massive inequality because it doesn't personally affect us as far as we know?

What I see is a lot of people terrified of simply releasing the data because the data will look, in many cases, very bad. I think if people want to advocate against any sort of intervention - perfectly reasonable to not want intervention - they should be able to support their case with data.

Now do prisons, the murderrate, etc.

Men live shorter lives than women do that must mean there’s systemic discrimination against men.

That makes no sense. You don't just hire people out of the whole general population, you hire people with skills you need. If a tech startup hires engineers those will be mostly male. Not because the startup is sexist, but because there are more male engineers around.

The united states has a population of ~332 million as of 2021 numbers. You're saying that a software startup's selection of available engineers is so limited that it's not possible to approach general population demographics?

In 2021 alone, over 100k people graduated with bachelor's degrees in computer science. That sample size is too small to approach general population demographics?

Okay, let's lower the bar. Why doesn't your staffing at least approximate the demographics of computer science graduates? Or even narrowing it further - let's say your startup needs the absolute best - people with PhDs in computer science? (According to the 2021 numbers, ~2500 people and 25% female)

You can argue over what the numbers should look like and how we should achieve those numbers but it's a cop-out to claim it's impossible or "too hard" due to a limited supply of hires.

> You're saying that a software startup's selection of available engineers is so limited that it's not possible to approach general population demographics?

Yes, because engineers don't follow general population demographics.

> In 2021 alone, over 100k people graduated with bachelor's degrees in computer science. That sample size is too small to approach general population demographics?

Yes. Again, that sample too does not follow population demographics.

> Okay, let's lower the bar. Why doesn't your staffing at least approximate the demographics of computer science graduates?

It does. Which is why I said "mostly" (but not exclusively) male. However that is not a "lowered bar", that is the only possible way things can work out. Unless you argue that a number of male engineers should be systematically doomed to never be hired in the name of equality, because otherwise we cannot reach your perfect demographic distribution.

https://www.bls.gov/iif/oshwc/cfoi/cfoi2006_13.pdf

92% of the 5840 occupational deaths in 2006 were men. Why isn't it closer to 50%?

A fantastic question! Someone should do something about it. Maybe you can?

Did you have something in mind?

The problem is selective enforcement. "your race/gender distribution should at least vaguely resemble that of the general population" Does this get applied to the NBA, benching the majority black teams and getting some 5'6 Asian or Hispanic males onto the court who Represent America?

Generals get promoted for having more women or people of color promoted under their command. When the casualty reports come in and white men are dying on the battlefield as disproportionate rate, double their demographics do we have generals saying "your race and gender distribution should at least vaguely resemble that of the general population" so we aren't going to send white men on the front lines anymore, we're going to get some diverse front line soldiers until our casualty rates look like America?

Reply on news.ycombinator.com