I saw a pretty impressive cheat tool that could apparently grab the screen from the live share, process text on the screen in response to an obscure keybind and then run it through OCR to solve (or just look up a LC solution).
At that point it seems like trying too hard, but be aware there are theoretical approaches which are extremely hard to detect (the inevitable evolution of sticky notes on the desk, or wall behind the monitor).
I’ve done the “at home” test for ML recently for a small AI consulting firm. It's a nice approach and got me to the next round, but the way the company evaluated it was to go through the questions and ask "fundamental ML bingo" questions. I don't think I had a single discussion about the company in the entire interview process. I was told up front "we probably won't get to the third question because it will take time to discuss theory for the first two".
If you're a company that does this, please dog food your problems and make sure the interview makes the effort feel valued. It also smells weird if you claim it's representative of a typical engineering discussion. We all know that consultancy is wrangling data, bad data and really bad data. If you're arguing over what optimiser we're choosing I'd say there's better ways to waste your customer's money.
On the other hand I like leetcode interviews. They're a nice equalizer and I do think getting good at them improves your coding skill. The point is to not ask ludicrously obscure hard problems that need tricks. I like the screen share + remote IDE. We used Code which was nice and they even had tests integrated so there wasn't the whiteboard pressure to get everything right in your head. You also know instantly if your solution works and it's a nice confidence if you get it first try, plus you can see how candidates would actually debug, etc.
I've seen it implemented in a couple of ways. One of the simplest is stronger blinding of applications so you aren't biased by irrelevant personal information.
For example this has been pretty effective in astronomy (proposals to get time on competitive instruments) where applications with female names were less likely to be approved. Simply removing the name was enough to improve the situation.
The fallacy is that people think DEI encourages preferential or token "diversity hires". When implemented correctly, it should mean that successful applications (for jobs, grants, whatever) go to the most qualified or deserving candidate using unbiased/impersonal metrics. That may mean encouraging minority groups to apply, but at the time of review, you should have no idea who the applicant is.
Unfortunately, the incorrect implementations are much more common in my experience. Out of the 4 companies I've worked at 3 carried out explicit discrimination under the banner of DEI.
> For example this has been pretty effective in astronomy (proposals to get time on competitive instruments) where applications with female names were less likely to be approved. Simply removing the name was enough to improve the situation
> The fallacy is that people think DEI encourages preferential or token "diversity hires".
Why do you say it is a fallacy when the E in DEI stands for the so called equity, which means exactly that. You seem to be describing equality, which is something that DEI was intended to supersede.
When you look at pre-Trump guidelines for FAA job applications, for example, there is an explicitly described preferential process for hiring from the favored groups.
Equity doesn't mean that though. There are people (like the current president) who certainly have something to gain from selling you that bridge, but no one I know who works in or advocates for "equity" initiatives considers them to advocate for token or diversity hires.
Have you considered that your sources of data may be biased?
Consider, for example that need-based scholarships are perhaps one of the clearest examples of a equitable, but explicitly unequal program. Should they be removed, or perhaps is there some legitimacy to this "equity" thing?
> Consider, for example that need-based scholarships are perhaps one of the clearest examples of a equitable, but explicitly unequal program. Should they be removed, or perhaps is there some legitimacy to this "equity" thing?
That's discrimination on the basis of income, which is legal. Discrimination on the basis of race and gender is not. All of the DEI programs I've seen have focused on equity with respect to race and gender. And 3 out of 4 companies I've worked at employed discrimination to that end.
Plenty of people have seen their employers adopt discriminatory practices in pursuit of equity. Condescending towards people and insisting that they're being misled about DEI is not a good way to defend it when loads of people have witnessed discrimination carried out under DEI programs firsthand. While I respect that none of your experiences with DEI involved discrimination, I urge you to meet other people who've had different experiences with the same respect.
I want to break down this post a bit, because this topic is often divisive and I'm aiming to be constructive. So, the question I asked, "Is there some legitimacy to this equity thing", and the question you chose to answer, which as best as I can tell is "Is racial discrimination legal?".
Those aren't the same question and I don't think conflating them is constructive. I'm not arguing that companies can do harmful things in the name of diversity. But the goal of equitable outcomes being worthwhile is independent from anyone's feelings about the legitimacy of any particular approach to meeting that goal, and that's what I asked about, and what you didn't reply to. I'd ask you to consider why.
Reading some of your other posts on the topic, you keep quoting a "3/4" statistic, and you've described different practices when asked to describe what they were you've described:
- Explicit racial and gender quotas in headcount (which are flatly illegal and so I doubt)
- Hiring goals in OKRS (which aren't inherently discriminatory, and probably weren't achieved!)
- Giving candidates multiple attempts on interviews in certain cases
- Requiring a certain number of underrepresented candidates to be offered interviews before the candidate can be picked
I want to dig into this last one, because in my experience, this is somewhat common, but only with executive hiring, where often roles aren't open for public application, and this forces executive search companies to make a modicum of additional effort to source and provide diverse candidates to the hiring company. This has, in some cases I've seen, also resulted in executive level positions having open public job postings.
I also want to dig in to something you said in another comment, because I think you're using discrimination in a particular way that's not how most people think of it. Phrased more forcefully, the definition of "racial discrimination" you appear to be using and considering harmful includes practices that are both generally considered legal and generally considered moral.
> Not every form of discrimination involves lowering hiring standards. For instance, imagine I flip a coin whenever a Catholic candidate applies. Tails, their resume goes into the garbage bin, heads and their application process as normal. Does this lower lower hiring standards for non Catholics? No. Does this advantage non-Catholics over Catholic candidates? Yes. It would halve the hiring rate of Catholics, though it doesn't result in any "lowering the bar".
So, I agree that this would lower the hiring rate of Catholics, but under this definition, ending this (clearly) discriminatory practice would lower the hiring rate of non-catholics. If your metric is "it is discriminatory if it harms the hiring rate of some group", then you've created an explicitly zero-sum definition of discrimination, and anything that reduces discrimination against some group causes discrimination against another. In doing so you've baked in an assumption about the correctness of the methodology at some particular time and place that you're measuring against.
You can call that discrimination, and I actually do in a lot of circumstances, but it isn't bad. It's actually a good thing, and we recognize that in a lot of circumstances that discrimination in pursuit of equity is moral and legal and even necessary!
Outside the US, for example, there are explicit racial quotas in a lot of situations, for example in New Zealand[0] there are explicit quotas in parliament to ensure continued native representation, and India has quota systems to help support lower caste individuals in opposition to active discrimination in society by upper-caste folks.
Even in the US, Due to the 1965 Voting Rights Act [1], ensuring representation of minority groups and actively "discriminating" in district drawing to ensure minority representation is both legal and in some cases required. And while the supreme court has defanged the voting rights act in recent years, it still recognizes that looking at race explicitly as part of disparate impact tests makes sense in many cases.
So while I do in fact meet with and talk with lots of people who have different experiences, I urge you to do the same: consider if your interpretation of what is harmful discrimination is in line with common use.
There is no zero sum of discrimination. In the above example, removing the coin-toss for Catholics removes the discrimination against Catholics, but does not add any discrimination against other groups.
Your example of India's caste based affirmative action is a good one. In fact, one of my past employers replicated it: we reserved a segment of engineering headcount for women, much in the same vein as India's reservation system with respect to caste. However, this is illegal in the US, but the company was confident that non-discrimination laws wouldn't be enforced when the victims were men. Though as of the last election, companies seem to recognize that this non-enforcement is a thing of the past.
The 3 out of 4 data point is my firsthand experience with my previous employer's DEI practices. Out of the 4 companies I've worked at, 3 employed DEI practices that are overtly discriminatory. So understand that when you write about how discriminatory DEI practices are myth, people such as myself have firsthand experience to the contrary. You're asking people to disbelieve their own lying eyes, when you insist that the notion that DEI involves discrimination is misinformation.
Equity requires discrimination, since almost no outcomes are perfectly equal across demographics. Achieving an equitable gender representation of pediatricians requires discriminating against female pediatricians. Achieving an equitable gender representation of software developers requires discrimination against male developers.
My use of the word "discrimination" is squarely in line with common use: does a policy discriminate on the basis of protected class? If so it's discrimination. Reserving headcount for one gender, even if it's done to push gender representation closer to 50/50, is discrimination. The notion that discrimination isn't discrimination when it's perpetrated against "non-diverse" groups is not in line with the common use of the term.
> There is no zero sum of discrimination. In the above example, removing the coin-toss for Catholics removes the discrimination against Catholics, but does not add any discrimination against other groups.
I want to clarify, because I see an inconsistency here. You say that preventing Catholic resumes from getting thrown out solely on the basis of race, is not discrimination and a good thing, and I agree. But elsewhere, you've said that changing hiring processes to ensure that URM and women are searched out and interviewed, is bad discrimination.
The same action, ensuring fair access to the pipeline, is in your opinion non-discriminatory (and legal) when it helps theoretical Catholic people, but is discriminatory and illegal when it benefits women and racial minorities.
I'm calling this out for two reasons, one because I want you to consider and reflect on this inconsistency, and two, to point this out to others reading and engaging, because insisting that we're asking you to disbelieve your own eyes does feel compelling, but less so when it appears very apparent that objectively, you're perception is colored by bias.
> So understand that when you write about how discriminatory DEI practices are myth, people such as myself have firsthand experience to the contrary.
I want to clarify, I've never said anything like this. My comments are pointing out an objection to your understanding of the word equity and how you interpret it. I've been fairly consistent about this, and you continue to avoid that and make more sweeping objections and statements about programs and your experiences, which may or may not be true but aren't actually relevant to the central question I'm asking, which is "Can equitable programs be good?"
> Equity requires discrimination, since almost no outcomes are perfectly equal across demographics.
I'm also going to call this out yet again, because it's the objection I made at the beginning of this thread. So I'll outline it yet again because you avoided it twice now:
I consider need based financial aid to be an equitable program, specifically because it takes into account the circumstances of the individuals. It does not require or make any assumptions about outcomes being equivalent. What it avoids doing is making the assumption that giving all people the same treatment (the same price) is the best approach. Do you disagree, and if so where?
Now I could agree that it requires discrimination based on financial status, but that is an input to the system, not an output (both in individual cases, and in aggregate). Where I disagree is that equity requires some assumption that outcomes are perfectly equal across demographics. I think that if you believe that, it makes perfect sense to oppose such programs, but I also believe that the people doing most equity work don't believe that, and it is primarily opponents of the programs that push such narratives.
> My use of the word "discrimination" is squarely in line with common use: does a policy discriminate on the basis of protected class?
You're using the word in its definition. And much like in this case, you've conflated what I'd describe as "unlawful discrimination" that violates the Civil Rights Act and "immoral discrimination" that may or may not break laws but is a bad thing, and "discrimination" that may in fact be a good thing but is predicated on recognizing differences between groups elsewhere in your responses, and it muddles the points that I think you're trying to make.
I'd agree that equitable programs require discrimination in that third sense, but I disagree that such discrimination is necessarily unlawful or immoral, and I'll yet again point to the example of need based financial aid as an explicitly legal, moral, equitable category of program.
There is no inconsistency at all. A policy that uses protected class as a factor, in any way, is discriminatory.
> But elsewhere, you've said that changing hiring processes to ensure that URM and women are searched out and interviewed, is bad discrimination.
"changing hiring processes to ensure that URM and women are searched out and interviewed" is a vague statement that encompasses both discriminatory and non-discriminatory practices.
Examples of non-discriminatory hiring practices that could boost women and URM participation include: anonymizing hiring processes to prevent bias or recruiting at events like Grace Hopper or at HBCUs. Examples of discriminatory practices include: Quotas (whether they're quotas on hiring or interviewing), reservation systems that prohibit non-diverse candidates from segments of headcount, or bonuses paid out to managers when they hire candidates of a particular race or gender. All of these practices serve to increase diverse turnout, but the former do so in a non discriminatory manner and the latter do so in a discriminatory manner. The problem is, only the latter can guarantee equitable outcomes.
You're using financial aid as a motte-and-baily here. Financial aid and preferences for low-income applicants in university are legal. Preferences for race and gender in hiring is not. Both are discrimination, yes. But the former is legal, and the latter is not. There's no inconsistency in supporting the former, and opposing the latter. And the conversation is about racial and gender equity.
> What it avoids doing is making the assumption that giving all people the same treatment (the same price) is the best approach
When it comes to race and gender in hiring, it is mandated by law that you give applicants of all races and all gender identities the same treatment. Anything less is illegal. I guess it's possible to think that the best approach is to flout the law and carry out illegal hiring practices to push towards equity. And at 3 out of the 4 companies I've worked at that's what they did, so you'd be hardly alone in that position.
> I'd agree that equitable programs require discrimination in that third sense, but I disagree that such discrimination is necessarily unlawful or immoral
Again, the conversation is about racial and gender equity in hiring. I don't care about your opinion on financial aid about preferences for low income college students. When you wrote this statement, "I disagree that such discrimination is necessarily unlawful or immoral", are you referring to discrimination on the basis of race and gender in hiring?
I highly suggest you read up on employment laws, especially since you're putting your employer's name in your profile.
> There is no inconsistency at all. A policy that uses protected class as a factor, in any way, is discriminatory.
But the policy you describe uses protected class as a factor, and results in, for example, atheist applicants being less likely to be hired than before.
> And the conversation is about racial and gender equity.
No, the conversation is about equity, and my belief that you misunderstood the term. All I've claimed is that equity is a worthwhile goal. You seem deeply unwilling to engage with that claim, instead distracting with questions about legality which, even if your understanding were correct, wouldn't be relevant to my question.
> Quotas (whether they're quotas on hiring or interviewing)
So would a policy that says that for any given position, you must interview at least one candidate from say 4 distinct declared racial backgrounds be legal in your view? It treats every race equally and doesn't favor or disfavor any particular race.
> I highly suggest you read up on employment laws, especially since you're putting your employer's name in your profile.
While the relevant executive order was repealed, making it no longer required, I think referencing the department of labor's own info on hiring law is worthwhile: https://www.dol.gov/agencies/ofccp/faqs/AAFAQs and https://www.dol.gov/agencies/ofccp/faqs/placement-goals#Q3 The crux being that while explicit hiring quotas are illegal, which is what I said, goals to measure against are not illegal. There's nothing I can find to suggest that quotas in application pool size are unlawful, and they're every common. I appreciate your concern, but perhaps read up yourself.
> But the policy you describe uses protected class as a factor, and results in, for example, atheist applicants being less likely to be hired than before.
No it does not. When orchestras put a veil between the auditioner and the evaluators and women's representation subsequently increased, that wasn't discrimination against men. That was discrimination against women being removed.
A policy of rejecting 50% of Catholics discriminates on the basis of religion. Removing this policy and adopting a non-discriminatory hiring practices may result in lower proportional representation of non-Catholics. That doesn't make it discriminatory, it just meant that the representation of non-Catholics was previously inflated by the discriminatory suppression of 50% of Catholic applicants. I'm not sure why this is hard to comprehend.
> So would a policy that says that for any given position, you must interview at least one candidate from say 4 distinct declared racial backgrounds be legal in your view?
Nope, this is still using protected class as a factor in hiring. You cannot deny, nor delay, the hiring process on the basis protected class. Meta used this policy, they called it "Diverse Slate Approach". I explain how it discriminates on the basis of protected class here: https://news.ycombinator.com/item?id=42753672
> While the relevant executive order was repealed, making it no longer required,
It was never even allowed by law. The executive order doesn't actually do anything except restating the protections of Title VII of the Civil Rights Act of 1964: discrimination on the basis of protected class is illegal in hiring. What it really did was signal that non-enforcement of antidiscrimination laws against "non-diverse" groups is no longer going to continue. This is why Meta ended the DSA, and why many institutions are rolling back their DEI programs.
> The crux being that while explicit hiring quotas are illegal, which is what I said, goals to measure against are not illegal.
Imagine one company tells its executives that their performance reviews will be penalized if their department fails to meet a 35% quota for women in engineering roles.
Another company tells its executives that their performance reviews will be penalized if their department fails to meet a 35% goal for women in engineering roles.
Whether you call them "goals" or "representation targets" or "diversity milestones", whenever you define success and failure in terms of numerical thresholds on the basis of protected class, it's almost certainly going to incentivize discrimination.
And again, in my experience companies are simply ignoring these laws and adopting quotas or other discriminatory practices anyway. Or at least they were during the previous administration, when they were confident enforcement would be lax.
> That doesn't make it discriminatory, it just meant that the representation of non-Catholics was previously inflated by the discriminatory suppression of 50% of Catholic applicants. I'm not sure why this is hard to comprehend.
No no, I fully agree with you here. I just also continue this train of thought to say that doing the same thing by removing discriminatory hiring practices that were suppressing the hiring of women and URM applicants is also reducing discrimination. But somehow you seem to consistently disagree at that point.
To build on your example, imagine that the company doesn't immediately realize that their hiring process is tossing out resumes. They think it's working fine. Someone comes along and says "hey, your hiring of Catholics is surprisingly low compared to what we would expect." The company investigates and agrees. They realize it's not actually due to the system throwing out 50% of Catholic's resumes, but instead some automated screening software that misinterprets language that Catholics use and throws out a disproportionate number of their applications. Unfortunately, this system is extremely effective in other cases, so the company can't actually just entirely stop using it, or they'd be unable to effectively screen candidates at all.
The company devises a plan, with a measurable goal to increase employment among Catholics by 10% by the end of the year. They re-review some Catholic resumes resumes, and proactively reach out to some applicants who had been denied and re-interview them. They put some additional systems in place to manually review new incoming resumes by Catholic applicants. They work with the automated resume screening software provider to improve the software and stop rejecting the applicants. They reach out to Loyola and Georgetown and ensure that the students at these traditionally Catholic universities have some additional training on how to write their resumes in a way that gets past the screening software.
This all seems fine and good, both morally and legally, to me. Where do you land?
> You cannot deny, nor delay,
> I explain how it discriminates on the basis of protected class here:
I don't actually see any justification for saying a delayed hiring process would be discriminatory. Unless you were to try and claim that the intent of the delay were to get non-diverse applicants to leave the hiring pool, there are pretty reasonable business justifications, such as "any hiring process that is unable to find even a single potentially qualified minority applicant is probably irrevocably flawed."
To zero in on your Meta example, unless you can show that the hiring process was noticeable longer for some candidates than others, which I find deeply unlikely, the more reasonable explanation than your handwavy suggestion that
> But the way it "improved" the demographics of Meta was through systematically delaying offers on the basis of protected class.
is that it improved demographics by exposing decision makers (Hiring Managers) to more qualified applicants from diverse backgrounds. To phrase this differently: if we return to our Catholic example, if you implement a requirement that all roles must consider a Catholic, and this encourages recruiters to actively reach out to catholic hires, some of the folks who were autorejected by the screening process will be found despite being missed the first time.
> It was never even allowed by law.
I want to clarify, you are saying that the Department of Labor guidance as defined under https://en.wikipedia.org/wiki/Executive_Order_11246, which has been the Department of Labor policy for 60 years now, is unlawful, and has been for 60 years, through multiple administrations on both sides of the aisle? Because that seems like it is beyond the pale of a reasonable understanding of the law. Of course it was (and still is) legal for companies to have diversity goals.
Given that the EO doesn't change the underlying law, while it is obviously no longer necessary for companies to comply, what they were doing before was clearly lawful and they can continue to do it!
> Imagine one company tells its executives that their performance reviews will be penalized if their department fails to meet a 35% quota for women in engineering roles.
Quota has a specific meaning in these contexts, that among other things includes the concept of including objectively less qualified folks on the basis of the discriminated attribute. So a gendered quota would mean something like hiring the top N applicants from each group, even if for example, there were more qualified men being passed over, and this was actively acknowledged. If it's not doing that, legally speaking, it isn't a quota.
> Another company tells its executives that their performance reviews will be penalized if their department fails to meet a 35% goal for women in engineering roles.
This requires deeper analysis: are they hiring only qualified applicants, but making new efforts to appeal and access previously untapped groups of qualified women? Great!
> This isn't a hypothetical example. Your own company created such a policy: https://www.reuters.com/technology/google-evaluate-executive... I wouldn't be surprised if Google does away with these totally-not-quotas the same way Meta ditched the DSA.
What Meta chooses to do or not is totally up to them, but they are not the arbiters of legality. This policy was implemented in the previous Trump admin. It was legal then. It's legal now.
> Whether you call them "goals" or "representation targets" or "diversity milestones", whenever you define success and failure in terms of numerical thresholds on the basis of protected class, it's almost certainly going to incentivize discrimination.
Put simply, I disagree. There are all sorts of ways to improve representation that aren't illegal. You seem to claim, however, that a number of commonly accepted, widely adopted, standard policies that companies have used for decades are actually illegal and have been the entire time. You understand why that causes me to doubt your reliability as an arbiter of what legal vs. illegal discrimination is, right?
Like, when you say "in my experience companies are simply ignoring these laws and adopting quotas or other discriminatory practices anyway", and I see the things you object to, which aren't quotas and aren't illegal, and have in some cases survived judicial review, and in some cases, are explicitly required by the government (see, for example, the consent decree in https://www.eeoc.gov/newsroom/radiant-services-pay-11-millio..., which requires "a recruitment plan and meet hiring goals to recruit, hire or place workers that reflect the percentage of non-Hispanic hires that would be expected based on the composition of the relevant labor pool"), you're saying that the actively practiced interpretation of the Title VII and the EEOC's practice for decades now is requiring illegal discrimination, and that's clearly untrue. You just disagree with the government on the specific kinds of discrimination that should be illegal, and that's fine, but you also shouldn't mislead people and claim that companies are doing things that are unlawful when they are just doing things you disprefer.
> I don't actually see any justification for saying a delayed hiring process would be discriminatory.
So I can institute a policy of delaying offers to pregnant women for 9 months?
> Unless you were to try and claim that the intent of the delay were to get non-diverse applicants to leave the hiring pool,
Yes, that is the intent. The longer candidates have to wait for an offer, the more likely they are to accept a competing offer or drop out of the hiring process. The way the DSA increased the representation of "diverse" demographics is by making "non-diverse" candidates wait a potentially long time for the DSA to be fulfilled, while diverse candidates never had to wait because their inclusion immediately fulfilled the DSA.
You realize that "increasing diversity" is identical to "reduce the representation of non-diverse demographics"? It's a discriminatory policy that examines applicants protected class and treated them different based on their race and gender.
> You seem to claim, however, that a number of commonly accepted, widely adopted, standard policies that companies have used for decades are actually illegal and have been the entire time.
These policies haven't been used for decades. Google only recently including representation quotas - oh, sorry, goals - in performance reviews in 2021. Meta didn't employ the DSA until 2018. These are have not been employed for decades.
> Quota has a specific meaning in these contexts, that among other things includes the concept of including objectively less qualified folks on the basis of the discriminated attribute. So a gendered quota would mean something like hiring the top N applicants from each group, even if for example, there were more qualified men being passed over, and this was actively acknowledged. If it's not doing that, legally speaking, it isn't a quota.
>> Another company tells its executives that their performance reviews will be penalized if their department fails to meet a 35% goal for women in engineering roles.
>This requires deeper analysis: are they hiring only qualified applicants, but making new efforts to appeal and access previously untapped groups of qualified women? Great!
You realize these are the exact same policy? If I penalize my employees if they don't meet a minimum percentage of a protected class, it's the same policy regardless of the specific word uses for the minimum percentage.
"Your performance review score is reduced if you don't meet a quota of X% women"
"Your performance review score is reduced if you don't meet a goal of X% women"
The incentive structure this puts on employees is exactly the same. At this point you're defending quotas, as long as companies use a euphemism for them.
I do agree that these quotas are in common use. But you're wrong that using a euphemism for quotas makes them any different in function.
Throughout your replies, you repeatedly defend the use of non-discriminatory policies that don't treat applicants differently on th basis of protected class, and only involve soliciting more applications from diverse candidates. That's all fine and dandy. But then you turn around and try to defend policies that do discriminate on the basis of race and gender. This doesn't follow. The fact that a quota could theoretically be fulfilled by more aggressively soliciting applicants from the target demographic doesn't change the fact that it's a discriminatory policy. Especially when the quota calls for levels of representation well above the demographics's representation in the field, it's certain to result in biased hiring. And no, simply calling a quota something else doesn't change that fact.
The motte and bailey you've constructed is clear:
The motte: companies should analyze their hiring policies and make sure that candidates are treated no different on the basis of protected class. Yes, I agree.
The Bailey: companies should discriminate on the basis of protected class, including practices like delaying hiring for non-diverse candidates and setting numerical thresholds on the basis of protected class (so long as a word other than "quota" used) is used for those thresholds. No, not only do I not agree, such practices are prohibited by law.
> So I can institute a policy of delaying offers to pregnant women for 9 months?
Then I think my caveat ("Unless you were to try and claim that the intent of the delay were to get non-diverse applicants to leave the hiring pool") applies.
> Yes, that is the intent. The longer candidates have to wait for an offer, the more likely they are to accept a competing offer or drop out of the hiring process. The way the DSA increased the representation of "diverse" demographics is by making "non-diverse" candidates wait a potentially long time for the DSA to be fulfilled, while diverse candidates never had to wait because their inclusion immediately fulfilled the DSA.
You can justify this numerically, right? Like, You have some data to back this up, right? Like you aren't just entirely making up a boogeyman, right? You have some evidence beyond your personal dissatisfaction and disapproval that the process actually results in significantly different processing times for diverse vs. non-diverse applicants, right? Because without such evidence, you're making a pretty conspiratorial claim.
> You realize that "increasing diversity" is identical to "reduce the representation of non-diverse demographics"? It's a discriminatory policy that examines applicants protected class and treated them different based on their race and gender.
Yes, I agree with you. But now when you said
> That doesn't make it discriminatory, it just meant that the representation of non-Catholics was previously inflated by the discriminatory suppression of 50% of Catholic applicants. I'm not sure why this is hard to comprehend.
You see how this is a contradiction. The policy to "increase diversity" and improve retention of Catholics isn't discriminatory, when you described it, but it is discriminatory now when it suits you since it "reduces the representation of <non-catholic> applicants".
You can't have it both ways, this is why the language you use is important, because while it is true that many of these policies do take into account applicant race and gender, they do so in ways that are explicitly legal, and often explicitly required. So when you say any policy which "examines applicants protected class and treated them different based on their race and gender" is illegal, you're wrong. That's not what the law says, that's not what the law thinks, that's not what the courts have held for 60 years.
If a company institutes a policy to reduce discrimination against a particular class, that policy must, necessarily recognize that class and consider it differently to ensure that it is treated fairly. The law recognizes this, you appear not to.
This is why I keep asking you to use more precise language, you're saying things are discriminatory, but you use that sometimes to mean in the illegal sense, and sometimes you mean in the practical sense of recognizing differences.
> You realize these are the exact same policy? If I penalize my employees if they don't meet a minimum percentage of a protected class, it's the same policy regardless of the specific word uses for the minimum percentage.
You continue to not understand what a quota is, and how the law considers them different than other policies. If you maintain the same hiring bar, it isn't a quota. A quota functionally requires using a lower quality bar for such applicants, and using class, not qualification, as the deciding factor.
> These policies haven't been used for decades. Google only recently including representation quotas - oh, sorry, goals - in performance reviews in 2021. Meta didn't employ the DSA until 2018. These are have not been employed for decades.
The policies I'm describing, that the EEOC has forced companies to follow as part of consent decrees, and that the federal government has required of contractors, have in fact been required for decades. That tech companies chose to adopt something similar of their own volition and then un-adopt a few years later does not change the underlying legality of the class of policy.
No one was requiring Meta to implement specific diversity policies, and so they were free to stop doing them. That's true, but also the policies they had were not illegal, and other companies have done similar things for longer, and some companies are required to institute similar policies and have been for longer than Meta has existed.
> setting numerical thresholds on the basis of protected class (so long as a word other than "quota" used) is used for those thresholds
I continue to believe companies are able to achieve diverse hiring practices without resorting to illegal discrimination. You seem unwilling to believe this is even possible in theory, and contend that the simple act of having a diversity goal is de facto illegal. That simply isn't true, and I'd really appreciate it if you could read the links I've posted, such as https://www.eeoc.gov/newsroom/radiant-services-pay-11-millio..., where the government requires companies to institute such policies. You understand why it is odd to contend that the EEOC's operating procedure for the past 60 years is, on its face, illegal? If it were, you'd think someone would have won a court case to that effect.
I'd once again appreciate if you tell me which of the policies I suggested that the company undertake to increase Catholic representation are illegal, because as far as I can tell, all of them are illegal, and I'd like to know where or if you disagree.
> You can justify this numerically, right? Like, You have some data to back this up, right? Like you aren't just entirely making up a boogeyman, right? You have some evidence beyond your personal dissatisfaction and disapproval that the process actually results in significantly different processing times for diverse vs. non-diverse applicants, right? Because without such evidence, you're making a pretty conspiratorial claim.
Yes, if you compile statistics on how long candidates have to wait for the DSA to be fulfilled, you'll see that diverse candidates don't have to wait at all because the DSA is fulfilled the moment that a diverse candidate is interviewed but the non-diverse candidates are stuck waiting until a diverse candidate is interviewed.
This is explicitly documented in Meta's hiring policies, both the mechanisms by which offers for non-diverse candidates are delayed and the goals of reducing non-diverse representation.
> You see how this is a contradiction. The policy to "increase diversity" and improve retention of Catholics isn't discriminatory, when you described it, but it is discriminatory now when it suits you since it "reduces the representation of <non-catholic> applicants".
The "The policy to 'increase diversity' and improve retention of Catholics" in the example above is to stop rejecting 50% of Catholic applicants.
If a company anonymizes its hiring and the hiring rate of women subsequently increases, then that is increasing the representation of women through non-discriminatory means.
If a company institutes a policy where managers' performance reviews are penalized unless they hire a certain percentage of women, then that is increasing representation through discriminatory means. The performance reviewed were changed to use the protected class of hires as a factor.
I'm not sure why you think this is a contradiction. If a policy uses the protected class of applicants or hires as a factor, it's discriminatory. Things like anonymizing interviews are not discriminatory. Do you really not understand why anonymizing hiring is non-discriminatory, but setting thresholds on the hiring outcomes and penalizing employees if those thresholds are not met is discrimination?
> You continue to not understand what a quota is, and how the law considers them different than other policies. If you maintain the same hiring bar, it isn't a quota. A quota functionally requires using a lower quality bar for such applicants, and using class, not qualification, as the deciding factor.
A quota does not require setting a different hiring bar. Imagine I have a diversity goal of 40% women hires in software development. Only 20% of software developers are women, so in order to meet this quota, I flip a coin whenever a male developer applies and reject 50% of male applicants. The rest of the interview process proceeds as normal. This doesn't lower the bar for women. They still need to pass the same set of skills based interviews as men. But it's absolutely discrimination on the basis of gender to reject 50% of male applicants to pursue a diversity goal.
"Your performance review score is reduced if you don't meet a quota of X% women"
"Your performance review score is reduced if you don't meet a goal of X% women"
Why does one require lowering the bar, but the other does not, when the policies are functionally identical? Your performance review score is reduced if you hire less than X% women in both cases. Is your line of argument really that the mere substitution of one word determines how employees will respond to the threat of lower performance reviews if a threshold isn't met? That's ridiculous, both policies exert the exact same amount of pressure to meet the X% threshold. If penalizing people for failing to meet an X% quota will lead them to lower hiring standards or engage in discrimination, then there's no reason to think that the exact same policy save for the word used to refer to the threshold would have any different outcome.
That was done as part of a settlement for a discrimination lawsuit. The exceptions to the prohibition on using race and gender as a factor in hiring are narrow. Bona fide occupational qualifications are one. You can specifically hire a Back actor to portray, say, Frederick Douglass in a movie. Consent decrees as part of a discrimination lawsuit are another.
But none of these apply to Google, or any of my past employers that engaged in discriminatory hiring. Your justification for discrimination on the basis of protected class only applies in a narrow set of circumstances.
> That was done as part of a settlement for a discrimination lawsuit. The exceptions to the prohibition on using race and gender as a factor in hiring are narrow. Bona fide occupational qualifications are one. You can specifically hire a Back actor to portray, say, Frederick Douglass in a movie. Consent decrees as part of a discrimination lawsuit are another.
So your contention is that the government regularly requires companies to undertake otherwise illegal activities, and that it would be illegal for the company to voluntarily follow a similar set of policies? Can you point to a statutory basis for this claim? I don't think I've ever seen the concept that a consent decree can require otherwise illegal actions.
> Why does one require lowering the bar, but the other does not, when the policies are functionally identical?
My contention is that it does not require illegal activities to meet reasonably set hiring goals.
> If a company institutes a policy where managers' performance reviews are penalized unless they hire a certain percentage of women, then that is increasing representation through discriminatory means. The performance reviewed were changed to use the protected class of hires as a factor.
In your Catholic example, is saying "we have a goal to stop flipping the coin" illegal? It is otherwise identical to a goal of increasing representation of a certain group by a certain percentage by changing some aspects of the hiring process that unfairly impacts a particular class.
Why is that policy acceptable, but real world policies are, seemingly in every case, supposedly illegal?
> This is explicitly documented in Meta's hiring policies, both the mechanisms by which offers for non-diverse candidates are delayed and the goals of reducing non-diverse representation.
Why haven't you sued them, if this was documented to the degree you describe, it was on its face illegal. And meta was actively admitting to illegal discrimination. I've been awarded money in a similar suit, so it's certainly possible for white men to win in discrimination cases.
> So your contention is that the government regularly requires companies to undertake otherwise illegal activities, and that it would be illegal for the company to voluntarily follow a similar set of policies?
Not regularly, only when companies are successfully sued for discriminating.
> Can you point to a statutory basis for this claim?
United Steelworkers vs. Weber established that use of protected class as a factor in employment is allowed only in narrow circumstances. In that instance, a company only hired steelworkers from a whites-only union, and were compelled by the government to give preferences to nonwhite employees for some period of time afterward.
In normal circumstances, a company cannot give racial preferences.
> My contention is that it does not require illegal activities to meet reasonably set hiring goals.
You still haven't answered the question:
"Your performance review score is reduced if you don't meet a quota of X% women"
"Your performance review score is reduced if you don't meet a goal of X% women"
Why does one require illegal activities to meet, but not the latter? This remains unanswered.
> In your Catholic example, is saying "we have a goal to stop flipping the coin" illegal?
It's not.
> It is otherwise identical to a goal of increasing representation of a certain group by a certain percentage by changing some aspects of the hiring process that unfairly impacts a particular class.
Because it's using protected class as a factor in hiring.
Say an orchestra adopts anonymous auditions, and the representation of women rises from 33% to 50%. That's a non-discriminatory hiring policy.
Another orchestra issues a "goal" or "representation target" or "diversity milestone" or some other euphemism that allocates 50% of the orchestra seats to women. That's a discriminatory hiring policy.
This is fourth time I've explained this, it seems like you're being willfully ignorant here.
> Why is that policy acceptable, but real world policies are, seemingly in every case, supposedly illegal?
They're not illegal in every case. Anonymizing interviews is legal, for example. The problem is that the legal ways of reducing discrimination don't lead to equitable results. Thus pushes towards equity require illegal discrimination, like setting thresholds and penalizing employees when those thresholds aren't met.
> Why haven't you sued them, if this was documented to the degree you describe, it was on its face illegal.
Because I'm not a victim of its policies. I'm Cuban so I'm one of the "diverse" candidates that fulfills the DSA. I have no basis to sue.
As phrased, I don't think the first would be illegal. It probably doesn't actually meet the definition of a quota, although use of the word "quota" is likely to receive stricter scrutiny.
If it can be met without illegal practices, it's perfectly fine, and if you don't actually institute a quota, even if you call it a quota, it's fine.
If the goal can be met by expanding the talent pool or by reducing discrimination, it doesn't matter what you call it.
> Say an orchestra adopts anonymous auditions, and the representation of women rises from 33% to 50%. That's a non-discriminatory hiring policy.
> Another orchestra issues a "goal" or "representation target" or "diversity milestone" or some other euphemism that allocates 50% of the orchestra seats to women. That's a discriminatory hiring policy.
Now what if they are the same orchestra, that is, they set a goal of hiring 50% women, and meet it by blinding auditions? Again, my contention is that it is not necessary to undertake illegal activities to achieve a diverse hiring goal.
You seem to agree with this in theory, but when given any concrete policy proposal, even ones that are widely accepted and have been adopted by wide swaths of the US for the better part of a century, you claim that they are illegal. You're simply incorrect.
> Not regularly, only when companies are successfully sued for discriminating.
This is approximately a weekly occurrence (they settle about 2 cases per week), I'd say that's fairly regular.
Further, such goals were required by all federal contractors from 1965 to Jan 20 of this year (https://en.m.wikipedia.org/wiki/Executive_Order_11246), its kind of silly to continue to claim that diversity goals are illegal when they've been required by every federal contractor for longer than any of the companies you've mentioned have existed.
> As phrased, I don't think the first would be illegal
"Your performance review score is reduced if you don't meet a quota of X% women" You think this is legal?
> Now what if they are the same orchestra, that is, they set a goal of hiring 50% women, and meet it by blinding auditions?
And what happens if they blind the auditions, but the share of women still falls short of 50%? Well, then they have to adjust the results of the blind auditions to match the quota.
You've repeatedly done this sneaky rhetorical slight of hand, where you endorse explicitly discriminatory policies and justify it with the claim that it could be achieved through non-discriminatory means. But that doesn't change the fact that the policy is discriminatory, and almost certainly would result in discrimination.
If I tell my employers that I don't want to hire any pregnant women - say, we institute a "diversity goal" of 100% non-pregnant employees - they could achieve this goal without actually discriminating against any pregnant women. That doesn't change the fact that such a goal is discriminatory.
> Again, my contention is that it is not necessary to undertake illegal activities to achieve a diverse hiring goal.
I've witnessed firsthand companies setting diversity "goals" of 40% women electrical engineers. 10% of electrical engineers are women. There's no way this can be feasibly achieved without resorting to gender discrimination.
> You seem to agree with this in theory, but when given any concrete policy proposal, even ones that are widely accepted and have been adopted by wide swaths of the US for the better part of a century, you claim that they are illegal. You're simply incorrect.
I've listed numerous hiring policies that I explain are not illegal:
* Anonymous resume review
* Adopting objective evaluation criteria
* Create and send resumes to your recruiters, identical save for identifying information like names, and follow up on any discrepancies on call-backs.
> This is approximately a weekly occurrence (they settle about 2 cases per week), I'd say that's fairly regular.
~100 cases a year nation wide makes it an extremely rare occurrence.
> Well, then they have to adjust the results of the blind auditions to match the quota.
No they don't, they can (and in cases I've seen!) fail to reach the goals! Or they can adopt other legal policies such as additional ones I've suggested, or they can change the goal. But in any case the mere presence of a goal is not evidence of illegal activity, even here you yourself show that the actual illegal activity wouldn't be having the goal, but doing some illegal activity to meet the goal if the legal methods weren't enough.
Which is to say, we agree: a diversity goal itself is not illegal. You should instead say that you believe that the diversity goals you are criticizing are too high to be met by legal means. That's potentially an interesting conversation, but its much less punchy than what you're saying now, which is functionally that "the mere act of committing to diversity in a quantifiable way is criminal".
> Your performance review score is reduced if you don't meet a quota of X% women" You think this is legal?
As long as the methods of increasing representation are legal, yes. You yourself have said as much. Consider a sufficiently low X, such as 1%. I don't think it would be illegal for a company to punish managers who hire > 50 employees per year for failing to hire even a single woman over the course of a year. I think that would be strong evidence of discrimination on the part of those managers, and the companies would be right to discipline or fire employees who engaged in such discrimination.
> You've repeatedly done this sneaky rhetorical slight of hand, where you endorse explicitly discriminatory policies and justify it with the claim that it could be achieved through non-discriminatory means.
No, I've been consistent that it is not illegal to have a hiring goal. Such a goal can be met in legal or illegal ways. You appear to consistently claim that even stating a goal and holding people accountable to it is illegal. I disagree. It depends entirely on what methods are chosen. The pursuit of diversity is not inherently illegal.
> I've listed numerous hiring policies that I explain are not illegal
And you've additionally claimed that a number of perfectly legal policies which are unlawfully discriminatory. Pipeline expansion policies with the aim of creating diverse candidate pool are perfectly legal, especially when they aim to account for flaws in existing talent pipelines that may be biased by missing underrepresented candidates.
Your objection to to these programs is that you claim meta announced that the goal of one of their programs was to delay giving interviews and offers to certain candidates with the intent of having those candidates drop out, so they could instead extend more offers to diverse candidates, and that you saw numerical documentation of these delays and their impact. And all I'm going to say so that's the most bullshit thing I've ever heard, because of it were in fact that well documented, it would be a slam dunk case and have been front page news.
The fact that no one else I've ever spoken to about such pipeline expansion and slate diversity programs has ever mentioned such a concern suggests that this "fact" you have is not as well documented or as clearly intentioned as you make it seem.
I'm going to disengage after this post, because we aren't getting anywhere, you're accusing me of various bits of rhetorical sleight of hand that I'm not doing, and you appear to be inventing more and more elaborate lies to prop up your objection to these programs. That's your right, but it's not something I want to entertain further.
> ~100 cases a year nation wide makes it an extremely rare occurrence.
But I'll note you ignored the part where I mentioned every federal contractor has done things you claim are unlawful for 60 years.
Newer Sony Walkmans have lossless support and some can also act as a USB DAC. Unfortunately the UIs are still terrible (no search, only oversensitive alphabet scroll?), proprietary cables and the device slowly re-indexes every boot. Organizing music hasn't improved since 2000 with some albums being split into 10 artists if they have collaborations. Still, at least we don't have to use Sonic Stage any more, mine has a uSD slot, and mass storage largely works. It's a shame Zune went nowhere, the HD was a wonderful device.
There is nothing that says you can't, but you have to convince the consulate that you did not have intent to immigrate when you applied/interviewed for the E3 and entered the US. The restriction is similar for the J1.
You can't land and immediately start applying for a green card, but if say you worked a 2-3 year contract and decided you wanted to stay, you could pursue a green card that you're eligible for. Most immigration firms that specialize in green cards will give you free advice on what the best route is (based on your background).
It would work on the ground, I believe the pilots (normally) had to get a fix before takeoff. You do need to see the sky without cloud cover, but spy satellites were less of a concern back then so less risk of being overflown during a daylight setup. The cameras are basically visible telescopes with very narrow fields of view and good baffling. Only a few stars are bright enough that you can sight off them, but it can be done. The device does a scan, so it's only accepting a small area on the sky and the initial fix can be sped up because you know where/when the aircraft is taking off. A lot of tricks to minimize the need for "plate solving", like knowing which direction the aircraft is pointing within some tolerance.
It wasn't exactly a simple instrument to use, and it relied on a ton of planned course information. You could also do a cold midair start after a power outage, but preflight would be much more preferable!
Some modern microwave telescopes like BICEP3 have an additional optical telescope for star pointing that are daylight-usable, but in summer you need to use a big baffle tube. The images are taken with a high sensitivity CCD camera and you can pick out brighter target stars surprisingly well in the images.
BICEP3 actually uses a >20 year old CCD camera with analog video output (BICEP Array uses newer cameras, with more modern sensors). Daytime star pointings are possible by using a low-pass filter to block visible light and take advantage of the sensitivity of CCD / CMOS sensors to the near infrared, where the daytime sky is more transparent, combined with baffling.
I would add it also uses an ancient analog TV for manual sighting in combination with the GUI for semi-auto centroiding. I always thought that was funny to see, but it seems to work well enough. Also, inserting that baffle is somewhat terrifying because it slots into a hole next to the main vacuum window and if you dropped it on the membrane, bad things would happen. Always fun to bump into Polies here :)
It depends what you mean by useful. On its own, all you're doing is taking pictures of the sky and figuring out where the camera was pointing (and its field of view). Where it's useful is calibrating the pointing direction of other systems. It's fun to try the software at home (there is a public web interface), you just need a camera that can take long enough exposures to see stars without too much noise.
One of the more "useful" backyard astronomy tasks that is achievable for a dedicated amateur is variable star observation (eg AAVSO), because many stars don't need huge telescopes to observe and it's very expensive for a big observatory to stare at a single patch of sky for weeks. Nowadays we have instruments like LSST which is basically designed for this sort of surveying, but public data are still useful. And you do need to know exactly where you're pointing, so either you do this manually by pointing at a bunch of target stars, or you can use a guide scope that solves the field for you.
With images taken at night, you can run the images through Astrometry.net, which is a blind astrometric solver and will provide you with RA / Dec for most images, as long as you have at least a dozen or two stars visible. The code compares asterisms formed by multiple stars to index files built from Gaia or other similar data. This is the technique that's used more frequently for microwave telescopes located where there's a normal diurnal cycle, e.g., CLASS. The smaller the field of the view, the higher the precision, but it also works fine with a camera with a zoom lens.
BICEP, however, is located at the South Pole on a moving ice sheet, requiring frequent updates to its pointing model, and has six months of continuous daylight, so daytime star pointing observations are required. This requires a different technique. Instead of looking at asterisms with multiple stars, the optical pointing telescope is pointed at a single star using an initial pointing model, the telescope pointing is adjusted until the star is centered, and the offset is recorded. This measurement process is repeated for the few dozen brightest stars, which acquires the data needed for refining the pointing model.
This is ubiquitous in baking at least. Also in confectionery where phase changes and structures are important (the canonical example being tempering). The extreme is probably Modernist cuisine.
You can look at the book "ratio" which presents a small number of standard recipes as proportions, with some hints for modification. I'd also recommend Lateral Cooking which describes recipes in terms of spectrums of ingredient variation or addition, usually starting with the simplest form. Finally there's a lot of interest in physics for coffee brewing, particularly pourover, but I'm somewhat skeptical of the rigour in that field and how much of it translates to better tasting cups.
Far as I can tell, Jonathan Gagné doesn't have a dedicated page on his website, it's hosted by Scott Rao. His blog does have a lot of interesting experimental work on the physics which led to the book. As I mentioned, I think this is an interesting academic piece which is at least supplemented by some genuine research. In practice, I feel like getting better pourover is 90% about finding beans that you like, buying a quality grinder and using them while fresh.
For confectionery, Chocolates and Confections (Greweling, Culinary Institute of America), 2013, ISBN 978-1-118-76467-1 is a fun book. It's quite pricey but you can pick up used copies now and again. It's a technical book and requires a lot of equipment that the average home cook doesn't have, but I would consider it fairly authoritative for looking up how chocolate things are made (and even discusses considerations for setting up a business). Most CIA books are pretty good on the practical side and they tend to be very exact with ingredients (almost always by weight).
I found a good channel called Chronova Engineering who made a few videos about machining the mechanisms by hand. They use a couple of off the shelf parts like tiny ball bearings for the demonstrator (instead of jewels), but almost everything else is hand-turned.
Basically any good SSD manufacturer is fine, but I've found that the enclosure controller support is flaky with Sonoma. Drives that appear instantly in Linux sometimes take ages to enumerate in OSX, and only since upgrading to Sonoma. Stick with APFS if you're only using it for Mac stuff.
I have 2-4TB drives from Samsung, WD and Kingston. All work fine and are ridiculously fast. My favourite enclosure is from DockCase for the diagnostic screen.
At that point it seems like trying too hard, but be aware there are theoretical approaches which are extremely hard to detect (the inevitable evolution of sticky notes on the desk, or wall behind the monitor).
reply