Hacker News Reader

Just hire based on skill / experience and don't not hire someone because they are a minority. Basic fair hiring goes a long way.

If you want more diverse candidates advertise in more diverse places.

To go through some cringe-inducing ordeal is not helpful. People want to be judged based on their skills not skin / gender.

You clearly don't understand Equity.

Oh, I think initial guidelines had that.

And then it was all tramped on by some insane incentive like a bonus being dependent on how many "diverse" people a particular person actually gets to hire. Or a stack ranking based on the same.

All it takes is just one perverse component in the mix.

> it was all tramped on by some insane incentive like a bonus being dependent on how many "diverse" people a particular person actually gets to hire

my company has done this and while it's not illegal to provide the incentives, I don't understand how they can expect it not to result in illegal hiring behavior (discrimination based on protected class) by those optimizing for those incentives.

> I don't understand how they can expect it not to result in illegal hiring behavior

Of course they expect it to result in illegal hiring behavior. They just also expect to always get away with it, because some technically illegal things are nonetheless tolerated by society.

How does that overcome systemic bias, eg prevalent poverty in many minority groups. The poor don’t have resources, wealth or contacts, to become a super coder/manager/whatever so they don’t make it into your applicant pool based on skill alone in the first place.

The cycle continues.

Breaking the cycle is hard. You could target low income children but that leaves everyone looking for jobs today in the cold, etc etc.

It’s a tough nut to crack but trying to fix historical bias isn’t necessarily a bad thing.

You need to fix that at the education level. Offering free top notch education can go a long way.

I agree but that ignores people trying to make it today and doesn’t create any “heroes” or mentors for those in education today.

Usually, you need to start super super early. Achievement gaps show up pretty early and only get wider. And I don’t think most companies have the stomach for broad-based improvements to education starting all the way from elementary school or even preschool, particularly when the payoff is going to be when the first wave of kids in preschool make it out of at least undergrad.

Of course, the superlocal, balkanized property tax system used to fund schools really doesn’t help things.

> Offering free top notch education can go a long way

I find it strange that people think you can just magically provide a top notch school and things magically correct themselves. How can you completely discount the impact of the parents involvement with their children's educational life?

Paraphasing JFK, "Ask not what your school can do for your kid's education, but what you can do for your kid's education".

Parents play a huge part in imparting good values in their children and also holding their children accountable for their performance in school. Furthermore, when some kids do poorly that rubs off on all the other kids by lowering the bar and setting bad examples. The parents set bad examples for their kids and then those kids set bad examples for other kids. It's turtles all the way down.

"If you want to see the poor remain poor, generation after generation, just keep the standards low in their schools and make excuses for their academic shortcomings and personal misbehavior. But please don't congratulate yourself on your compassion." – Thomas Sowell

The systemic bias I faced was that people assumed I was poor, didn't have resources, wealth or contacts because I am a Native American.

My recently experience with the Choctaw Nation would have led me to think the exact opposite. Their new headquarters is absolutely amazing. Beautiful new buildings, and huge casino expansion underway too. They seem to have been taking care of the tribe members quite well.

But that may be an inaccurate assumption on my part as well.

It is impossible to solve systemic bias by using discrimination. It has never worked in the past despite an untold number of attempts. In a very similar way, violence create violence in very predictable cycles.

Like violence, the solution to systemic bias are simple but unattractive. Don't display and rub symbols in people faces. Do not create ambiguous borders between groups of people. Treat each person as individual and avoid generalizations. Add social and economical support based on lack of those resources on an individual level, from the bottom up.

The progress towards equally would look very different if people weren't constantly doing the opposite in the belief that this time the symbols will convince the enemy of the wrong doing, the ambiguous borders will work, the generalizations will send the important message, and that top to bottom approach will create a foundation for change.

Despite declarations that busing to desegregate schools failed in the 1970s and 1980s, that era actually saw significant improvement in educational equity. When the National Assessment of Educational Progress began in the early 1970s, there was a 53-point gap in reading scores between black and white 17-year-olds. That chasm narrowed to 20 points by 1988. During that time, every region of the country except the Northeast saw steady gains in school integration. In the South in 1968, 78 percent of black children attended schools with almost exclusively minority students; by 1988, only 24 percent did. In the West during that period, the figure declined from 51 percent to 29 percent. But since 1988, when education policy shifted away from desegregation efforts, the reading test score gap has grown — to 26 points in 2012 — with segregated schooling increasing in every region of the country.

¯\_(ツ)_/¯

The case is interesting, through I would compare it to a similar but failed experiment in Sweden where students of got free points in their grade if they applied to a program where there would be a minority. Why did one have some success and the other had none? I would go back to the things I wrote above about what to do and what not to do.

Providing free buses to people who live far away from desired schools do not rub the act in the face of people who live near those schools and do not need buses. It might cause a negative side effect that school buses becomes a symbolic proxy for being poor, making those intended to use them less likely to agree to use them, but it extends the option available for those who only have few options to begin with. In contrast, free grade points is much more displayed directly in the face of other fellow students, both to those who get into the program and those who were rejected with similar grades.

Free buses does not provide any ambiguous borders or rules once the student is at school. The requirements and expectations of the student is same as any other student. There are few situation where uncertainty and doubt can occur between buss and non-buss student, except maybe for times of heavy snow or road congestion. To my knowledge, when students who depend on buses can't reach the school and thus given a pass to skip school, the other students are given the same pass. If we go back to the Swedish example, requirements and expectations get ambiguous when people get in through other means than merit, causing tensions between those who got in through grades and those who got in through the free points system.

Both systems has generalizations, so I would say that a way to improve the desegregation efforts would have been to skip the qualification of race and simply provide the busing to any individual that is below a set amount of wealth and live a distance far away from the desired school in order to need a bus. Such change might have saved it from being abandoned in 1988.

For a majority individual to 'feel ok' with a minority jumping the line they must fundamentally believe their merits don't matter as much as the minority's group identity. Likewise, the minority must believe their merits don't matter either, only their group identity matters. Therein lies the problem, 'color blindness' and 'affirmative action' are diametrically opposed. The harder you push for the one is a push away from the other. Which is 'harder' for a society to learn - their merits don't matter or their group identity doesn't matter (in the context of say getting a job)?

It's such a shame, because we want as a society the hiring manager to go to the reservation job fair looking for applicants. But the manager knows that there are precious few 'qualified' applicants and we recognize they will tend to maximize the efficiency of their time.

Of course, it isn't an accident there are few qualified applicants in those places. Poverty has a holistic effect on a group. My hope is that groups can somehow separate financial poverty from poverty of ambition to try to break these cycles and make progress. Every 'group culture' has a unique ethos, many poor groups had rich ambition and rose out of their poverty over time. To me this is the only way I can see real progress being made without giving up on meritocracy (and by extension technological/scientific progression).

So the question becomes - what does it look like to cultivate group ambition?

1. Generational support: believe your situation is not guaranteed for your children. Fight for them. Sacrifice for them.

2. Education/medical: we pour resources into certain underprivileged places (not groups, physical locations). This isn't going against meritocracy, it's a long-term investment.

3. Recognize that human civilization is at its finest when people believe they can rise above their circumstances, despite all the external and internal factors pulling them down.

> Add social and economical support based on lack of those resources on an individual level, from the bottom up

How do you apply this to a system that has favoured certain groups for so long that the whole structure has close to none of those it supposedly wants to help?

You need representation from groups that need help in order to help. You need representative individuals in the organisation to guide it. You may have to positively discriminate at first, or else you won’t have any representative to draw from.

There are abundant examples of butchered attempts at ‘helping’ when those with good intent have no idea (this thread gives plenty of cringe inducing examples).

The key, I think, it making it clear what the role is and what is wanted to those that are there in that role.

There are disparities companies can fix through hiring practices and ones they can’t. Systemic bias is one they can’t—who Google hires or doesn’t simply doesn’t move the needle on generational poverty. Even if every professional industry does it, that doesn’t move the needle because those industries employ a small fraction of overall workers. Fixing biases and barriers to promotion in service jobs and trades would do vastly more to fix systemic bias, because those industries simply employ vastly more people across a bigger cross section of the work force.

Then there are disparities that better hiring practices can fix, namely correcting the “first and only” problem. Most people just want to fit in, and being the “first and only X” makes that harder. It’s a kind of disparity that exists only because it exists and can be fixed by simply hiring people from diverse backgrounds to achieve a critical mass so that future applicants don’t face that conundrum.

> who Google hires or doesn’t simply doesn’t move the needle on generational poverty.

If it can’t be done in one of the most successful companies on earth, in the richest country, who can?

The value in Google trying is greater than just the value to the successful applicants. It might not be much but just throwing up your hands because it’s hard is not helpful.

Why do you single out poverty and bias in minority groups? What about say, Appalachia? You think the white people living there were born with a silver spoon in their mouth?

Why someone is poor today shouldn’t be rated. It’s not a suffering rating contest. You could be denied a job 50 years ago because of your skin, your accent, etc.

It’s far better to just look at where someone is financially and use to determine if they need help. It also prevents divides from popping up. A very good way to get people angry is to tell someone who needs help they are privileged and should shut up because they are either white, male, straight, don’t look poor, etc.

Ideally, yes. But the problem is that our implicit biases affect how we view potential candidates.

Certain groups need the chance to prove themselves. If certain groups aren't fairly represented, then it does pay to prefer to hire from those groups when everything else is roughly equal.

I do find it hard to believe that "the best" candidate is more often than not a straight white man.

> I do find it hard to believe that "the best" candidate is more often than not a straight white man.

So do I, which is why I find this diversity hire exercise entirely pointless.

I think the point is that IBM believed they were hiring based purely on skill and experience. Then they realized they weren't hiring any native americans. And they figured that the best way to address that would be to hire some.

> this time they have a Native American employee, you, that they can talk to about how to reach out to that group.

I remember reading someone's opinion basically stating that things like that for underrepresented groups are unpaid emotional labor, which I kind of agree with - it's got to be a weird position to be in to be hired to write code and somehow end up as the defacto sounding board for questions regarding the company's interactions with people similar to you. Imagine having randos come up to you and ask you for your time to pitch new ideas for how to recruit white dudes or asking questions about how it was growing up a white dude or even just people making the assumption that since you're a white dude you care about hiring more white dudes.

> what other process would we suggest?

Personal opinion alert: I feel like a lot of (not all) tech's diversity problem is a pipeline problem so maybe identifying people that belong to underrepresented groups out there doing good work getting folks from the group interested in cs/tech and then giving them more resources?

> Imagine having randos come up to you and ask you for your time to pitch new ideas for how to recruit white dudes or asking questions about how it was growing up a white dude or even just people making the assumption that since you're a white dude you care about hiring more white dudes.

I imagine this all the time, yet it never happens.

> things like that for underrepresented groups are unpaid emotional labor

Presumably if the person in question is salaried and it's understood that they are spending work time on this, then it's not unpaid emotional labor. But, sure, it's work that requires leaning on emotional and social sensitivies one might otherwise not have to bring to bear in their tech job.

I don't think anyone should be obligated to do that kind of work because of their background. But my experience is that organizations and culture often progress only when some people do volunteer to do that work. And, fortunately, there are many people who choose to do that because the intangible rewards, for them, outweigh the costs.

This is not how things happen, I have yet to see any large organization that let itself or its procedures changed by new hires. And telling someone that was mistreated by a company that they should have accepted and then tried to change the things from the inside is insensitivity at its best. Now if they hire you specifically for that purpose that's different, but it abaolutely doesnt sound like it here.

You are wondering about the wrong thing. Companies/Orgs are not designed to pull this type of stuff off. And thats the end of the story right there.

Society has developed other mechanisms precisely because orgs are dumb as shit at handling complexity.

Its sort of like encouraging 2nd graders to think about 10th grade problems. They need a whole lot of help and time from someone else before they can. Thats why the East India Company runs to King or Goldman runs to the President or Jack Ma evaporates from the earth when things get too complex.

Orgs are just dumb simple parts of society like the ribosome is a part of a cell. Expecting the ribosome to morph into something it is not just because every part of the ribosome has good intentions and feels it can be doing much more to influence the behavior of one cell, or a clump of them and therefore the whole body makes for a good Pixar movie and Google hiring material. But in real life that stuff happens on very different time scales and by very different institutions.

If you want to change the world don't work in the corporate world. Work in places that supervise them.

Why would a Native American feel prospect feel more comfortable talking to a Native American employee? I think that's the cringey stuff parent comment is complaining about.

Greater number of shared experiences and common ground? Why do any pair of humans feel more comfortable talking to each other than some other pairs?

I think tech events focused on specific groups can be great, but from my perspective:

- Better if it is not the corporation organizing it, but just sponsoring either trough financial contributions or things like providing space instead.

- Have interesting topics beyond just talking about group X.

- Don’t assume everyone within a group is the same.

- Actual hiring/evaluation process should be same for everyone, do as much blindly as you can.

I always thought the best way was to treat those recruiting events as a way to get a more diverse set of candidates on the resume pile. Once they are on the same pile as everyone else, the higher ratios of minorities on the piles will naturally lead to more minority hires (assuming your hiring process is non discriminatory).