4 ms·
If there is systemic discrimination in favour of group A, then the only way to fix that discrimination is to eliminate it. To the members of group A (in this ca
by halostatue 9y ago
If there is systemic discrimination in favour of group A, then the only way to fix that discrimination is to eliminate it. To the members of group A (in this case, mainly white men), that feels like discrimination. It isn’t discrimination, but that’s what the loss of power and representation feels like. Even if you step it down through affirmative action programs, group A feels like it’s being discriminated against…
…which is only true if removing of unearned privileges is discrimination. (It isn’t.)
By “bar”, I mean that most hiring practices are crap. There’s a bunch of nonsense talked through on a whiteboard that has nothing to do with day-to-day programming or anything you’ll find on a job…and white men tend to do exceedingly well at that sort of crap. There’s an overrepresentation of specific skills required to pass the “technical” interview that are equally underrepresented in the technical execution of day-to-day work.
People pretend that it’s skills-based, but it’s skills-based with a bias toward people like yourself. What I’m saying is that in my own hiring practice, I do have a technical exercise. But it’s not one that requires that a developer “complete” it successfully, because the folks who measure the success are not looking for amazing technical solutions… they are members of the team who you’d be working with. They are assessing whether you listen to technical instruction; if you know more, how you express yourself to someone junior to you; if you know as much, how well you work with them. It’s a soft-skills developer test, because if you can demonstrate the ability to work through this problem with the other people in there with you—even if you don’t complete the test, you will probably fit on the team.
And the reality is that in ~40 interviews over the last two years, the real jerks have been white men (maybe two interviews total), including one who questioned the value of the technical exercise that I developed. (It’s valuable in part because I’ve had to implement the exact exercise for a job once because a language runtime did not have the data structure I needed.)
If you’re only measuring on what you think is a meritocratic skills-based solution, you are supporting the systematic discrimination against women and minorities, because they are always judged as less capable than white men.
- m52go 9y agoIt's hard for me to follow your argument. > By “bar”, I mean that most hiring practices are crap. Agreed. > People pretend that it’s skills-based, but it’s skills-based with a bias toward people like yourself. Presumably, you mean the bias is toward white males. But then you go on to say your preferred method is a soft-skills developer test that the team who'd be working with the hire evaluates. Great--that's how hiring should be done anyway. > And the reality is that in ~40 interviews over the last two years, the real jerks have been white men (maybe two interviews total), including one who questioned the value of the technical exercise that I developed. Okay, if that's how it worked out for you, great. The whole point of any hiring process is to determine which people will be the best fit for your needs. If your method does this, then it's a good method. > If you’re only measuring on what you think is a meritocratic skills-based solution, you are supporting the systematic discrimination against women and minorities, because they are always judged as less capable than white men. This is where your argument breaks down. You started out talking about unearned privileges, but your whole argument is about how you test for desired skills, and then somehow in the end, you make it about race again (but this time as it relates to meritocratic hiring, not unearned privilege). So...where does race fit in to your hiring process? To me, that part of it just sounds like pandering, since your actual process sounds like it's all skill-based anyway.
- hardwaresofton 9y agoBasically just said the same things -- see my response. I think what they were trying to get across was that BECAUSE of the broken testing, tests that were supposed to be meritocratic are just checking for certain groupthink. My response there was if that's the case, the problem is just making better (actually strictly meritocratic) tests.
- hardwaresofton 9y agoI agree with your point that systemic discrimination is a problem, but I'm not sure that affirmative action programs are as simple as "stepping it down". I get that as things normalize group A will feel like they're being discriminated against, but I think that's precisely the reason affirmative action programs are controversial -- they're hard to distinguish from just discrimination in the other direction. Maybe that's the only option available, that discussion is long and nuanced. I also agree that most hiring practices are crap, and that interviews don't properly reflect day-to-day work. I also agree with a point (I think) you made that a bunch of the "meritocratic" tests that are developed today are a lot of the time just checking if one was groomed in the same way (learned about algorithms the same way, learned the same practices, etc) and not necessarily pure can-you-get-shit-done tests. Your point gets a little hazy towards the bottom, for me, I can't tell if you're against supposedly meritocratic tests altogether and suggesting that people test for team-fit more. Your last sentence seems to have that same off-putting insinuation, just crystalized, but I think I understand your point, maybe I can rephrase it: "If you're measuring a candidate with what you think is a meritocratic skills based tests, it's easy to be blinded by inherent bias and support systematic discrimination against women and minorities, because many seemingly meritocratic skills based tests have more to do with confirming shared experiences/biases than checking actual skill." If I did indeed read your point well, my response would be to conduct better interviews, and create an actually meritocratic interview process, the onus is on the company.
- halostatue 9y agoAffirmative action programs are controversial because people who have privilege generally don’t like other people getting that privilege (because then they are no longer special). As an analogous situation, men often claim that women talk too much. There’s fairly strong evidence that the claim is merely sexist nonsense[1]. So forgive me if I’m unwilling to give a crap about whether people who look like me (white men) feel like they’re getting the short end of the stick because there are people like me working to end the special privileges that they have gotten literally for decades. Especially since so many of them really aren’t all that. The London School of Economics makes it clear that gender quotas actually improve meritocracy by squeezing out mediocre men[2]. Your rephrasing is mostly correct, but I would go so far as to say that if you don’t know how to see your own biases, you cannot design a meritocratic interview process that doesn’t inherently have bias against women and/or minorities. As this entire discussion on the firing of the author of the screed shows…there’s a lot of people who can’t see that they have biases. (And yes, I have biases. I have to work very hard to prevent those biases from causing me to discriminate. This is largely why I involve my team in hiring and constantly discuss the problems with bias blindness with them, too.) [1] http://thejunglenook.tumblr.com/post/75517868909/do-women-talk-too-much-hint-science-says-no http://thejunglenook.tumblr.com/post/75517868909/do-women-ta... … but also see [3] as an extension because it actually gets to the heart of just why the (now-ex) Googler at the top of this was just so wrong. [2] http://blogs.lse.ac.uk/businessreview/2017/03/13/gender-quotas-and-the-crisis-of-the-mediocre-man/ http://blogs.lse.ac.uk/businessreview/2017/03/13/gender-quot... [3] http://allthingslinguistic.com/post/145374253955/do-women-talk-too-much-hint-science-says-no http://allthingslinguistic.com/post/145374253955/do-women-ta...