@tikimixologist's banner p

tikimixologist


				

				

				
0 followers   follows 0 users  
joined 2022 September 04 23:09:57 UTC
Verified Email

				

User ID: 257

tikimixologist


				
				
				

				
0 followers   follows 0 users   joined 2022 September 04 23:09:57 UTC

					

No bio...


					

User ID: 257

Verified Email

Do they want it? Clearly they were crazy about Black Panther. Black Panther is, aside from being a better-than-average and more imaginative capeshit title, a coherent movie inherently valorizing black people.

I will strongly dispute the idea that Black Panther is coherent. Various pieces from the movie: (spoiler warning!)

  • As we have seen in real life, being a semi-hereditary monarchy on top of natural resources leads to a nation skilled in science and technology.

  • Villain: "Ok Mohammed Bin Salman, you've defeated me in this battle to rule our Kingdom and I'm about to die. Here's a historical reference to stuff that happened in Brazil 150 years ago."

  • The central conflict of the movie is about Trumpian isolationism vs Clintonian internationalism. Black Panther starts the movie rescuing some Congolese women from child soldiers wishing to (presumably sexually) enslave them. But he's unmoved and still wants to build the wall. Then he changes his mind after hearing what life was like in Oakland 1992 (not, you know, Rwanda 1994, one country over from Wakanda) and becomes an interventionist.

It purports to be take place in a foreign country, but the entire country is nothing but vague ideas that American writers saw on the History Channel. For example, it's Africa and they watched a documentary about the Maasai in 1850, so modern soldiers should look like this: https://en.wikipedia.org/wiki/Black_Panther_(film)#/media/File:Dora_Milaje_in_film.jpg Also they watched Animal planet, thought Rhinos were cool and noticed they live in Kenya, so unobtanium doesn't just power technologies like clean energy and lasers but also improved animal husbandry.

This is not a coherent movie. It purports to tell a story about Wakanda, but every single plot line is driven by characters caring more about Americans of the same race as them than their own corner of the world. (And by "same race", I mean US Govt defined race as opposed to Bantu/Nilotic/Pygmy/etc. )

The academics doing AI ethics tend to be doing simpler things, e.g. trying to figure out what a "fair" lending algorithm is.

The technical challenge is finding an algo which spots hidden patterns that predict loan repayment except for the biggest pattern that predicts repayment (namely that blacks are much less likely to repay them, holding all else equal). But stating it in such explicit terms is a cancellable offense.

For you, sure. But it might not be for others - such as me - since among other things, I don't think I'm the same race as you.

I guess you didn't read my links carefully. The specific quote I provided of a woman trying to prevent his presidency was from an inauguration day protest (i.e. before Trump assumed the presidency) at literally the location where he would be inaugurated. Violent actions - e.g. setting a car on fire - also happened. So by your stated criteria, it was a coup attempt.

But I guess you can gerrymander your definitions even more carefully now that I've pointed this out.

The protests against Bush in 2000 and 2004 were also pre-inauguration, and were generally aimed at influencing the vote counting process.

it was literally a protest in the sense that nothing they could do at this point could make Trump a not-President and they were just expressing their frustration.

You seem to be claiming that because anti-Trump protesters (including violent ones) had no hope at achieving their stated goal of preventing him from becoming president, they are "just expressing their frustration". But when anti-Biden protesters (mostly peaceful) engaged in protest but had no hope of stopping Biden, it's a coup attempt. Weird.

Based on the research I did when my wife was pregnant, I came away with the understanding that both this overall fact and the specific causes were fairly well known: gestational diabetes and preeclampsia.

Gestational diabetes is not suspected to be directly racially linked, rather the suspected pathway is black -> (pre-pregnancy obesity, during-pregnancy weight gain) -> gestational diabetes.

Preeclampsia is less understood, but is known to be highly correlated to sickle cell and is theorized to be influenced by sickle cell's recessive form as well. Fibroids (main cause: low vitamin D which has the literal biochemical cause of more melanin -> less vitamin D) are also suspected to play a role.

In any case, if we wanted to actually measure racism as opposed to blame everything for it, the simple way to do so would be to look at the underlying rates of gestational diabetes and preeclampsia. Treatment after getting these conditions is susceptible to racism but actually getting the conditions is not. So high rates of (preeclampsia death / # of people with preeclampsia) might be evidence of racism, but (# of people with preeclampsia / # of pregnancies) would be evidence of bad genetic luck.

Phonics and DI aren't "silver bullets". They are merely interventions that reliably perform better than all known alternatives. Their inability to solve all problems is not a reason to give them up.

Also teaching outcomes are quite straightforward to measure, thanks to standardized tests.

From my first comment:

data strongly supports the position of educated internet right wing racists.

What more would you like?

If you actually look at nuclear development, electricity deregulation made it impossible to do the long-term funding to build nuclear reactors, because the time to get your money back is such a long tail.

Depends on whether you look at the cost before or after the government imposed regulations that make it impossible for nuclear to be cheap, specifically the "as safe as possible" standard (as opposed to "meet X bar of safety as cheap as possible").

https://postimg.cc/PLQH3hdn

It's perhaps worth contemplating who was president at the time of the price spike.

Tokyo is not stagnant in terms of population.

https://viz.wtf/post/158158642063/tokyo-population-over-the-years-look-at-what (Yes, the graph is bad, but it's the only one I can find with years after 2010.)

Since about 2000, Japan is (on a per-capita basis) no more stagnant than the US. If Tokyo were full of NIMBYs, people could afford to pay more.

https://i0.wp.com/fabiusmaximus.com/wp-content/uploads/2019/01/Per-Capita-Real-GDP-of-US-and-Japan.png?resize=889%2C400&ssl=1

If you want to find a metric on which Tokyo and major US cities/metro areas differ, try housing units built. In Tokyo, this was about 100k units/year since 1998.

https://www.nippon.com/en/japan-data/h00782/

So why don't landlords increase prices to capture more of that GDP growth? Because Tokyo has nearly 1M vacant apartments and if a landlord increases prices too much their flat will join the vacant pool. Simple as that.

At peak housing bubble, the entire state of California managed 150k houses/year and as of 2016 it was about 50k.

https://journal.firsttuesday.us/wp-content/uploads/California-Annual-Construction-2017.png

As I mentioned in another thread, we could build 19M homes in Santa Clara alone if we simply increased the density to that of San Francisco. We just choose not to.

Why can’t we make a giant metropolis in Oklahoma? And build it to 30 million people.

Or better - why can't we make a giant metropolis in Silicon Valley? The current population density of the San Jose-San Francisco-Oakland, CA Combined Statistical Area) is 1.1k/sq mi and it has 7.7M people. The density of Brooklyn is 38k/sq mi and San Francisco is 18k/sq mi.

If we increased the density of Silicon Valley to that of Brooklyn, we could fit 266M people there. We could fit 21M people alone in Santa Clara (San Jose plus a couple of suburbs) if we increased the density to San Francisco or Cambridge levels.

If you want to do it on unoccupied land, there is literally unoccupied land the size of San Jose directly adjacent to San Jose.

There are many places one can disagree with you on empirics. The most notable is here:

Diminishing marginal utility. At a certain point, another yacht for the ultrawealthy rich guy is not going to make him significantly happier.

The marginal alternate use case for the resources is investment in future production, not yachts. The question whether resources should be devoted to providing an x-box for poor people or to building electric cars/installing heat pumps/building homes/etc.

Moreover, this argument just sort of assumes resources are available and their quantity isn't affected by our choices. But in reality, the poor people are both consumers of utility and producers of it. The actual choice we need to make is between:

  1. A person refusing to work, being given resources anyway, and a marginal house is not inhabitable because no one is available to install drywall.

  2. A person installing drywall in return for a similar quantity of resources, but now we have an extra house that someone can live in.

It is far from clear that (2) is worse than (1).

Being afraid of falling into poverty is also bad for people's wellbeing -- it is a major source of worry and concern because everyone knows that being impoverished sucks and is painful. So the existence of poverty is a cause of pain for a much larger group than those actually impoverished. Fear of poverty also leads people to refuse to take risks to avoid the pain of poverty, which leads to less pleasure.

This is interesting. Possibly we should more widely publicize exactly what it means to live in poverty in the US? I.e. make sure everyone knows that "poverty" by US standards means lots of leisure time (most poor people don't work and aren't in the labor force), no danger of hunger, free medical care, a bigger house than the average Parisian or Londoner, 1-2 cars, etc.

From what I can tell, the only thing that's particularly bad about being poor in the US is that you spend time around other poor people.

Of course, knowing these facts does take a lot of wind out of the sails of the typical leftist who wants moar wealth transfers.

First is because of the Republican's insistence on 'Whiteness & Christianity' being core to the platform.

This is simply false and has been false for decades. Here's what George W. Bush said about the topic on Sept 17, 2001 for example:

The face of terror is not the true faith of Islam. That's not what Islam is all about. Islam is peace. These terrorists don't represent peace. They represent evil and war."

Leftists insist that Republicans are racist, and push this meme by misleading media stories. A Bush II example of this was James Byrd (black guy) getting murdered by racists in Texas while Bush was governor. Leftists who oppose the death penalty wanted a hate crime law, Bush said such a law was unnecessary because Texas has strict laws against murder. Texas eventually executed the killers based on those laws.

A Trump example of this is creatively editing a statement to imply Trump described white supremacists as "fine people" when he explicitly said he wasn't referring to them. (Full quotes here: https://www.politifact.com/article/2019/apr/26/context-trumps-very-fine-people-both-sides-remarks/ )

Now I fully believe that Indians and Chinese believe this to be true - but their belief has nothing to do with any actual mainstream Republican views. Your second reason explains quite well why Indian and Chinese Americans believe this, regardless of how true it is.

There have been similar protests that included a small amount of illegal behavior every time a Republican has been elected president since 2000.

https://wgntv.com/news/hundreds-of-peaceful-trump-protests-overshadowed-by-violent-acts-arrests/

Some of these anti-Republican protests were even organized by foreign powers: https://thehill.com/policy/technology/358025-thousands-attended-protest-organized-by-russians-on-facebook/

Did those people not, in some way, think their actions might lead to Trump being deposed? Their explicit statements to the media suggest they too were engaging in a coup attempt:

“Trump is illegal,” she said. “He is in violation of the constitution. I am doing everything I can to prevent his presidency.”

People don't typically use the term "anti-bias" to reference fixing bias in the statistical sense. It nearly always means preventing an AI from making correct hate-fact predictions or generating disparate outcomes based on accurate data.

Examples:

  • Lending algos/scores (e.g. FICO) are usually statistically biased in favor of blacks and against Asians - as in, a black person with a FICO of X is a worse credit risk than an Asian person with the same FICO. This is treated as "biased" against blacks because blacks tend to have lower FICO scores.

  • COMPAS, a recidivism prediction algo, correctly predicted that "guy with 3 violent and 2-nonviolent priors is a high recidivism risk, girl who shoplifted once isn't". That's "biased" because blacks disproportionately have a lot more violent priors. (There's also a mild statistical bias in favor of blacks, similar to the previous example.)

  • Language models which correctly predict the % of women in a given profession (specifically, "carpenter" has high male implied gender, "nurse" high female implied gender, and this accurately predicts % of women in these fields as per BLS data) are considered "biased" because of that accurate prediction.

(Can provide citations when I'm not on my phone.)

All of the examples you describe are simply examples of "making more accurate predictions", and that is totally not what the AI bias field is about.

The phrase "Narrow Bank" is the name of an actual bank that the Fed shut down because it did not lend out money and just held 100% safe reserves.

https://archive.is/TqCJX

Regional banks do have proprietary advantage - domain specific knowledge of regional industry and real estate as well as local relationships. JPM or BofA need to have fairly uniform underwriting rules and these rules will not necessarily allow them to service some particular industry with weird cash flow patterns (such as startups). Rapidly scaling SAAS is a very good example of this.

Another great service SVB provided is a degree of self dealing/moral hazard - founders can get a mortgage backed by illiquid equity if they do corporate banking with SVB. This sort of self dealing is a problem for startups, but it is not really a problem and is a useful feature for family offices.

Being too lazy to look up dates on Wikipedia is doing even more work. His first acts of ecoterrorism (arson, spiking trees) were in 1975. His first bombing was May 25, 1978. The incident with the limericks happened in Aug 1978.

https://en.wikipedia.org/wiki/Ted_Kaczynski#Initial_bombings

I suspect sex would be reasonably safe. But we already have a preview of what might happen if your personal utopia does is not what the zeitgeist wants. There was a Rimworld mod called "European Phenotype and Names Only (White Humans)" which modifies a single player game. It's banned.

https://www.eurogamer.net/paradox-pulls-discriminatory-stellaris-mod-that-made-all-humans-white

We embrace the idea that players mod the game to best represent how they want to play, we do NOT however wish to enable discriminatory practices.

I think that the term 'government funded media' clearly takes on a negative implication that extends beyond the strict meaning of the words.

This is of course the point. The American establishment uses terms like this as epithets to characterize their opponents, and simply avoids using these terms - even when literally true - for it's own mouthpieces. Now twitter is simply using these terms literally and in an unbiased manner and the establishment is losing their minds.

When you're accustomed to privilege, equality feels like oppression.

(By "unbiased", I simply mean twitter seems to be applying a standard of "does their website say they get X% of their money from the government", at least based on Musk's tweets.)

For every year they can't pay that 20k down at 5-8% interest, it's another 1k minimum added on top. A high debt(>60k) maximum repayment period is 30 years

At an 8% interest rate (charged only to grad students) and 30 years of repayments, that's $146/month. ($167 at 20 years, and $242 at 10 years.) Undergrads pay 5%, which means it's $107/month. If college doesn't increase your income by $107-146/month post-tax, you shouldn't go.

https://www.bankrate.com/loans/student-loans/current-interest-rates/#best-rates

This isn't just raw bigotry - a person who's learned to do honor culture sparring instead of acting 'reasonably' and signaling disapproval subtlely, who hasn't practiced doing paperwork or writing essays, who doesn't have a support network of other rich people to fall back on, will in practice be at a disadvantage in life, even if they're just as 'biologically' smart as a rich white person.

There is at least one company that attempts to teach intelligent but culturally backward people to do upper class signaling: Bloom School, formerly Lambda. They quite explicitly teach lower class people how to perform class signaling - an example Austin Allred (the founder) gave is that they tell everyone to get a bank account before getting hired, since "just cut me a check" signals low class. They internalize the benefits by charging customers a fraction of their paycheck assuming they get a job that pays at least $50k/year.

It's a popular target of dishonest hit pieces by left wing journalists, near as I can tell for exactly this reason. Journalists are high class, Bloom/Lambda students are low class, and yet Bloom graduates earn more than journalism school graduates.

As for the "why do Jews and Asians succeed then" question...When you're excluded from good occupations by law, your culture will steer away from them and towards unproductive activity or crime,

Every western state of the Union except Washington explicitly included Asians in Jim Crow laws. Southern states did not, probably because the vast majority of Asians lived in the west.

Montana: "Negroes, Chinese and Japanese"

Arizona: "Negro, Mongolian, Malay, or Hindu"

California, quite famously the location of lots of anti-Asian discrimination: "Negroes, mulattos, Mongolians and Malays"

Nevada was more descriptive about what these terms mean: "Ethiopian or black race, Malay or brown race, Mongolian or yellow race, or Indian or red race"

Oregon: "Negro, Chinese, or any person having one-quarter or more Negro, Chinese or kanaka blood, or any person having more than one-half Indian blood." (Kanaka = pacific islander.)

Utah: "white and Negro, Malayan, mulatto, quadroon, or octoroon void."

Insofar as Asians were not explicitly mentioned in Jim Crow laws the courts generally would include them should the matter be tested. For example in California the law specified "no black, mulatto person, or Indian" and the California Supreme court interpreted it to include Chinese: "It can hardly be supposed that any Legislature would . . exclud[e] domestic negroes and Indians, . . . and turn loose upon the community the more degraded tribes of the same species, who have nothing in common with us."

https://en.wikipedia.org//wiki/List_of_Jim_Crow_law_examples_by_state

I don't understand this complaint about the death penalty specifically. The alternative to the death penalty in most cases is life imprisonment which barely seems better. Why are false positives with life imprisonment ok, but the false death penalties not?

Moreover, life imprisonment lets us pretend that we'll fix the false positives later on, even though I see little evidence we actually will.

Overall this argument seems like an isolated demand for rigor.

Here's some some dueling anecdotes, provided from a hit-and-miss piece against Elon Musk:

They [Elon Musk and his siblings] had to make calculations when they were going out with nonwhite friends about what they could safely do, he said.

One time at lunch, a white student used an anti-Black slur, and Mr. Musk chided the student, but then got bullied for doing so, Mr. Ranwashe said.

Mr. Mashudu [Elon's BFF] was killed in a car accident in 1987, and Mr. Ranwashe said he remembered Mr. [Elon] Musk being one of only a handful of white people who attended the funeral in the family’s rural village.

“It was unheard of during that time,” he said.

Original version: https://archive.is/nmjHu

Updated version which takes most of the sting out: https://archive.is/26gMt

I call it "hit-and-miss" because the original version was clearly trying to make Elon Musk bad and racist for refusing to censor twitter, but failed horribly because the incidental anecdotes reveal Musk hating everything about apartheid including it's censorship regime.

If there were a number of minority business applicants who felt looked over in hiring and not adequately promoted, why wouldn’t they want to form their own business?

It's amusing that Goldman Sachs is your example of companies that currently hire Korean PhDs, given that Goldman was originally a company that hired Jews in a competitive marketplace where Jews were discriminated against. At that time Goldman mostly specialized in back office low visibility work - if the Jews get the numbers right, it's fine.

Also in common knowledge that has been memory holed, various popular left wing policies such as minimum wage and Davis Bacon were created to prevent a "race to the bottom" that resulted in greedy businesses hiring negros for cheap.

The general argument against this goes back to Becker, and it relies on a principal/agent problem. Consider a discriminatory firm which is driven out of business by competitors hiring cheaper negros. The former employees of this failed business do not exit the marketplace. Instead, they become employed by other businesses and continue to be racist. This racism may involve things like funneling the best sales opportunities to other white employees, shirking work when on a team project with negros, that kind of thing. The net result is that these racist employees drive down the productivity of black workers. That's the Beckerian theory.

(I use the term "negro" throughout much of this post to emphasize to the reader the time period I'm discussing.)

There's also the issue of network effects which I think is more modern. E.g. if your customers or suppliers are racist, a non-racist profit maximizing employer may put black employees in less visible positions. Literally all of the modern examples of this that I can think of result in discrimination against political conservatives:

  • Cloud hosting and other SAAS providers (e.g. Cloudflare) refusing to service conservative businesses (kiwifarms, parler).

  • Companies that "own the consumer" (Android store/Apple store) blocking access to consumers for conservative businesses (parler, veritas).

  • Passive shareholders funneling non-specific demand for investment to companies that engage in performances leftwing stuff (ESG). A famous example is giving Exxon a higher ESG rating than Tesla. "Indexing is communism" is the economic argument against indexing, but also "Indexing is globohomo."

  • Employees conspiring against political conservative coworkers, along the lines of Becker's theory. (I have personally witnessed an attempt by employees at getting a conservative fired, which luckily failed due to rigid company policies.)

Cost to launch a payload into space:

Space shuttle (USA): $54k/kg

Ariane 5 (Eurozone): $9.1k/kg

Proton (Russian): $4.3k/kg

Falcon 9 (Musk): $2.7k/kg.

Falcon Heavy (Musk): $1.4k/kg

https://en.wikipedia.org/wiki/Space_launch_market_competition

What a fraud. He definitely lacks the ability to execute.