Beware the Pretty People • Hacker News Board

Re: Beware the Pretty People

#101

Post by kbenson » Sun, Mar 01, 2015, 7:44 PM UTC

sago wrote: post #79

kbenson wrote: post #75

Earlier quoted context omitted.

It isn't very useful for getting to the bottom of what is actually happening. I disagree. Finding a shared vocabulary can be essential to actual communication. It's far too easy for people to argue a similar or even the same point without that shared reference point. That doesn't really need to be what it "really means" as long as it's meaning in the discussion is well understood. 'Merit' has its own problems, becaus…

> I disagree. I suspect you don't, because I agree with your counter-argument. I think language is flexible. It only matters that someone can understand what we mean, because it is the meaning that matters, not the word. Too often discussions end up about what words really mean or should mean. That feels like an argument that is getting somewhere, but it actually does very little intellectual work. It distracts from…

I suspect you don't, because I agree with your counter-argument.

Case in point.

Not to me. There's nothing gambling about, trying to make it so that potential capability is more accurately correlated to imminent capability. Then both strategies align.

I wasn't stating the choice is obvious, I was stating that I think what we currently do is obvious. The difference is do we hire based on potential capability, which there's no way of knowing for sure, or hire based on imminent capability, which is at least somewhat testable?

So we end up employing large numbers of sub-optimal people just because they could afford to become educated / mentored / inspired, and ignoring others with far more potential because they didn't happen to have those benefits.

Yes, and my argument is that I'm not convinced that an alternate method at the hiring stage would be beneficial in the long run. I believe we continue to hire those people that are sub-optimal in potential capability yet have higher imminent capability (through privilege, yes). To hire others would in no way guarantee a better outcome, even if you could identify the ones with untapped potential - there's no guarantee that potential could be developed well in many work environments. If done one a wide scale, there would be a necessary drop in societal imminent capability, but the real question is how fast does it recover (if it does), and does it result in a better eventual outcome than another strategy?

That other strategy I would propose is to hire as we currently do (mostly based on imminent capability), and leverage the gains we know this gives us into reducing the effects of privilege where possible. I think the United States has practiced this in greater or lesser degrees over the last century, and I think it's worked quite well. I just think changing the formula of who works based on capability is messing far too closely with one of the fundamental levers of capitalism, and there are less radical approaches that can and should be followed through on first.

Re: Beware the Pretty People

#102

Post by rezendi » Sun, Mar 01, 2015, 8:11 PM UTC

mc32 wrote: post #100

flycaliguy wrote: post #97

Earlier quoted context omitted.

"Accept that things which become mainstream have to acquire aspects of the mainstream." I guess my take away was that when weirdness becomes mainstream it becomes nothing more than a toothless affectation. That you should beware of the savvy newcomers who have learned how to adopt a style formed out of rebellion but will use it to push traditional objectives. They will push out the rule breakers while co-opting their…

You hit the nail on the head. Some people understand that for something to succeed, it has to become mainstream, but there are others for whom the original otherness _is_ their identity. It separates them from larger society, so when larger society takes on those signifiers, they feel threatened. Like the fashion cycle. This article preys on people's weaknesses --like feeling that their values or lifestyles are being…

"tech crunch," believe it or not, is not a hive mind with a single view point. (OP here.)

Re: Beware the Pretty People

#103

Post by mc32 » Sun, Mar 01, 2015, 9:27 PM UTC

rezendi wrote: post #102

mc32 wrote: post #100

Earlier quoted context omitted.

You hit the nail on the head. Some people understand that for something to succeed, it has to become mainstream, but there are others for whom the original otherness _is_ their identity. It separates them from larger society, so when larger society takes on those signifiers, they feel threatened. Like the fashion cycle. This article preys on people's weaknesses --like feeling that their values or lifestyles are being…

"tech crunch," believe it or not, is not a hive mind with a single view point. (OP here.)

I hope the irony isn't lost when tech crunch is one of the organizations which is not too different from the so called interlopers it decries. In other words, if not for tech becoming mainstream and significant tech crunch would not have much interest in reporting about tech. so it's not that they care about tech, they care about tech insofar as it affords them economic viability.

Do you see a 'tech crunch' of the mortuary industry? If embalming became the next big industry affording millions good earnings and made millionaires out of many, you'd see the same thing. And you'd see publications making much ado about nothing for reasons of self interest. Some times some of the things they would raise would raise valid questions but for the most part those publications would just ride the wave and decry the "posers" coming into the industry who didn't have to put up with decades of irrelevance and stigma --taking advantage of people's fears and weaknesses.

Just let people try and make a living wherever they see fit. Sure, they bring 'disruption' to the old guard, but so what. That's life.

Re: Beware the Pretty People

#104

Post by meric » Mon, Mar 02, 2015, 12:08 AM UTC

O____________O wrote: post #98

meric wrote: post #62

Earlier quoted context omitted.

There are plenty of non-white people in github. If there are still too many white people on github, you're free to analyse other programmers from websites other than github, such as this Chinese one: https://code.csdn.net/dwzteam/dwz_jui All "colours" of people have free time to create projects.

There are plenty of non-white people in github. I'd like to see how you analyzed the race of github contributors. I'd also like to know what "plenty" means. Personally, I strongly suspect that you'll see fairly small set of race/location/income combinations. All "colours" of people have free time to create projects. Given that we're not really talking about color, per se, I seriously hope that you don't honestly beli…

I'd like to see how you analyzed the race of github contributors. I'd also like to know what "plenty" means. Personally, I strongly suspect that you'll see fairly small set of race/location/income combinations.

Why, me and my friends and colleagues, Chinese and South African and Ghanian and Kenyan are on github. I'm very offended you suggested only white people can afford to be on github.

Given that we're not really talking about color, per se, I seriously hope that you don't honestly believe that statement.

You specifically mentioned "white males".

It takes half a day to write something you can put on github. If anyone can't do that it's because they're not as experienced as those who can. It's not much more effort than writing a resume. The point of github is so people who aren't interested in brain teasers can demonstrate their ability to write good code.

You can argue people who aren't good at programming enough to be able to write something in half a day should be hired anyway, and I'll tell you the communism experiment is over.

You, sir, are racially biased.

Re: Beware the Pretty People

#105

Post by sago » Mon, Mar 02, 2015, 3:15 AM UTC

proveanegative wrote: post #85

sago wrote: post #69

Earlier quoted context omitted.

When 'privilege' is used in these topics, it is often rather broadly defined to mean any feature of the accident of your birth that is beneficial compared to others who lack it. So yes, a long history of intellectual achievement is a privilege in that sense: the prevalence of expectation (you can do this), of value (it is worth doing this), of role-models (so and so, who is similar to you, did this), mentorship (so a…

>'privilege' is used in these topics, it is often rather broadly defined to mean any feature of the accident of your birth that is beneficial compared to others who lack it. Is that meaningfully different from "luck"?

From an individual's point of view, no.

But when privilege is correlated with a particular group, then it isn't helpful to see it as random. It self perpetuates.

But an individual can be lucky to be born into a rich family, but the rich family probably isn't lucky to be rich: it tends to retain its status over generations. Not in terms of absolute dollar value, but its social status.

Re: Beware the Pretty People

#106

Post by sago » Mon, Mar 02, 2015, 3:20 AM UTC

kbenson wrote: post #101

sago wrote: post #79

Earlier quoted context omitted.

> I disagree. I suspect you don't, because I agree with your counter-argument. I think language is flexible. It only matters that someone can understand what we mean, because it is the meaning that matters, not the word. Too often discussions end up about what words really mean or should mean. That feels like an argument that is getting somewhere, but it actually does very little intellectual work. It distracts from…

I suspect you don't, because I agree with your counter-argument. Case in point. Not to me. There's nothing gambling about, trying to make it so that potential capability is more accurately correlated to imminent capability. Then both strategies align. I wasn't stating the choice is obvious, I was stating that I think what we currently do is obvious. The difference is do we hire based on potential capability, which th…

I'm not sure.

When I was hiring for good C++ game engine programmers, our recruitment test had people program in a language they'd never seen before. Because I believed that the imminent capability of a person in C++ wasn't as good a determiner of their long term value as their potential capability, which I thought was more likely correlated to flexibility, learning speed, calmness under difficult conditions, clarity of reasoning.

So I think this already happens in hiring. What I'm describing is one end of a continuum. My point is that it isn't clear to me that day 1 competence is a good optimisation strategy. And if I optimize for day 10 competence, then why not day 100, 500, and so on? There's got to be a tipping point somewhere, but it isn't obvious where.

Re: Beware the Pretty People

#107

Post by sago » Mon, Mar 02, 2015, 3:27 AM UTC

yummyfajitas wrote: post #86

sago wrote: post #78

Earlier quoted context omitted.

Why are you so exorcised by the word privilege? My point was nothing to do with what you seem to have a problem with. My point was that what is seen as 'merit' is a function of a whole bunch of other factors, which means it isn't somehow a racists discussion stopping thing to point out that Asian men have higher levels of 'merit' (when merit is defined in the way I understood it to be in the post I was responding to)…

I don't know why you think I'm "exorcised" by a word. I'm simply pointing out that you are using a completely different definition of it than myself and hackeruser, and as a result your posts are completely missing the point. If privilege means "probabilistic causes of non-equal representation", then that's a very broad definition and every non-deterministic unequal representation is caused by privilege. But you were…

Didn't I specifically list some of the things I was talking about, explicitly?

Privilege is used in various senses: both the narrow power one group has over another, propagated by their ability to pass it to others in the group and more generally (in the later camp I gave the example somewhere on here of privileges of high IQ). I was trying to be clear in my OP by listing what I meant, so I really don't see why the word is what you're hanging on. OR why you think my purpose was to define it. As I said, call it boondoggle, I was explicit what I meant by it.

By definition, Asian males have 8x the "probabilistic causes of non-equal representation". It seems highly unlikely that they have 8x the "social biases unfairly favoring one group", which is what hackeruser and myself were actually discussing.

Why would the factors have to be equal for the point to stand? If something isn't 100% the cause, then it isn't significant? If it turns out that 4x the effect is due to cultural privilege, of the kinds I listed, and 4x is due to demographic, bias, so what?

Re: Beware the Pretty People

#108

Post by sago » Mon, Mar 02, 2015, 3:32 AM UTC

yummyfajitas wrote: post #88

marcus_holmes wrote: post #82

Earlier quoted context omitted.

hang on, but where are you drawing the lines between "merit due to innate skill/aptitude" and "merit due to favourable environmental factors"? Because similarly if you define merit as "anything that causes one person to do better in the tech industry than another" then the tech industry is tautologically a meritocracy - water is wet. Until and unless we can unpack merit and privilege in the manner that Sago is attemp…

I'm defining "merit" as the ability to get the job done and make the company money. For example, I'm trying to hire a data scientist/data oriented generalist. Knowledge of statistics and aptitude for learning contributes to merit, as does the ability to write scalable and deployable code. Sex appeal does not contribute to merit, nor does being a nerdy-but-attractive hipster. Does that clarify? Note that Sago's defini…

You do like ignoring that I actually gave a list, don't you:

money, aspiration, parental ambition, education, culturation, in-group bias, and many other factors

(note, many was not intended to be exclusive)

that you continue to think my point was to apply the label "privilege" seems obtuse, bordering on the willfully ignorant, since I've explicitly said that was not the point at all, and that the label is irrelevant.

Re: Beware the Pretty People

#109

Post by sago » Mon, Mar 02, 2015, 3:48 AM UTC

kbenson wrote: post #101

sago wrote: post #79

Earlier quoted context omitted.

> I disagree. I suspect you don't, because I agree with your counter-argument. I think language is flexible. It only matters that someone can understand what we mean, because it is the meaning that matters, not the word. Too often discussions end up about what words really mean or should mean. That feels like an argument that is getting somewhere, but it actually does very little intellectual work. It distracts from…

I suspect you don't, because I agree with your counter-argument. Case in point. Not to me. There's nothing gambling about, trying to make it so that potential capability is more accurately correlated to imminent capability. Then both strategies align. I wasn't stating the choice is obvious, I was stating that I think what we currently do is obvious. The difference is do we hire based on potential capability, which th…

I think it's worked quite well

Can you say what systemic biases you think have been driven out so far by doing this? I'm curious what you're thinking about.

I'd certainly say that there have been some improvements. But 'quite well' seems overly optimistic to me.

Re: Beware the Pretty People

#110

Post by yummyfajitas » Mon, Mar 02, 2015, 4:27 AM UTC

sago wrote: post #107

yummyfajitas wrote: post #86

Earlier quoted context omitted.

I don't know why you think I'm "exorcised" by a word. I'm simply pointing out that you are using a completely different definition of it than myself and hackeruser, and as a result your posts are completely missing the point. If privilege means "probabilistic causes of non-equal representation", then that's a very broad definition and every non-deterministic unequal representation is caused by privilege. But you were…

Didn't I specifically list some of the things I was talking about, explicitly? Privilege is used in various senses: both the narrow power one group has over another, propagated by their ability to pass it to others in the group and more generally (in the later camp I gave the example somewhere on here of privileges of high IQ). I was trying to be clear in my OP by listing what I meant, so I really don't see why the w…

You did not specifically list the things you were talking about - you ended the list with "and many other factors". That's not explicit at all. Then you said Call it 'probabilistic causes of non-equal representation', at which point I thought you were finally being clear. My mistake.

If it turns out that 4x the effect is due to cultural privilege, of the kinds I listed, and 4x is due to demographic, bias, so what?

Scroll up - as I said, you completely lost the context of the conversation.

Hackuser was asking "if a system is truly meritocratic then why would 38% of a population[1] produce 90%[2] of the winners?" The implicit assumption here is that "money, power, privilege" (as defined by hackuser) and merit are distinct, otherwise the question makes little sense.

Now if only 4x the effect is due to cultural privilege, then the answer to his question is "yes" - a perfectly meritocratic system can have disproportionate representation of Asian males.