@forest generally it's only used like that because of the right who bastardized it. before if you was anti-woke, complaining about issues that apply outside of the country was just a show of ignorance. eg gas prices increasing because of a war or manufacturing issues caused by global supply chain issues

"School district to lay off white teachers before minority teachers under new union contract." Really wish people would read past the headlines on this.

There's a reason the union negotiated this contract and it actually makes a lot of sense if anyone would care to actually pay attention.

It's lying headlines that make all the money though. :P

Like that one some months ago about the women who was kicked out of school for being raped... And it turned out it had nothing whatsoever to do with that, but "Christian school doesn't believe rape victim" is a better line to get people to click than anything else.

It's not lying though. In case of layoffs, the school district will prioritize laying off white teachers first, but there's a reason for it.

s/lying/misleading/ then

@Yuuki my first thought would be "because there are way way more white teachers"

Awhile back, the school district laid off a bunch of teachers, disproportionately black and minority teachers.

Obviously, this is bad.

what's the article anyway?

So they hired more minority teachers but the previous union contract, like most union contracts, used seniority as the metric for layoffs. If they kept using seniority, they'd just run into the old problem again because the minority teachers were least senior.

@Memor-X abcnews4.com/news/nation-world/…

As a compromise, they're using this race-based layoff priority scheme in order to maintain diversity in teachers.

I miss the days when the progressive ideal was colorblindness.

Back when equality as a goal was egalitarian and not identity-driven.

The problem is that meritocracy isn't race-blind. Or at least, it can't be in our current society.

Meritocracy is race blind. People are not.

Meritocracy is egalitarian if the culture and society are already egalitarian. But otherwise, it favors those with generational advantages, which tend to be non-minority currently.

The problem is people making decisions based on merit, which many refuse to do.

It's not just that.

Meritocracy is an end-goal, not a process. The end-goal is good, but it requires overcoming hurdles such as biases and bigotry. Making decisions with bias is very much anti-meritocratic.

Multiple socioeconomic studies have shown that your socioeconomic status as a child is a large predictor of your educational performance. It's not just "people are racist, that's why we can't hire based on skill". There's systemic and institutional racism that has nothing to do with people's perceptions and biases.

The meritocratic thing to do would be to try to equalize socioeconomic statuses.

E.g. extra funding in low-income schools is meritocratic, because you're empowering people.

Not to mention merit-based layoffs is easily manipulated by companies to be anti-union.

Then it's not merit-based. It's selfish, using merit as a shield in hopes that no one notices their actions.

e.g. dump all the work and set impossible metrics for pro-union employees

They can't complete the work, so they get written up, fired, and now you've weakened the union.

You're describing the antithesis of meritocracy.

You're describing a bastardization of it, meritocracy in name only corrupted by the greedy.

It's almost like corporations will bastardize anything to pursue control over their workers.

Nah, that can't be right.

Yup.

It's like saying taxing the rich is bad because the super-rich get around it anyway and it ends up hitting the middle class harder. When the super-rich are getting away with money laundering, it's not taxing the rich, it's a bastardization of the idea.

Likewise, if someone is setting impossible goals or specifically hiring people in such a way that they can claim merit-based layoffs that target minorities, it's not meritocracy.

Well, how do you prevent that? Because years of lawsuits will favor corporations with deep pockets.

Same way you make the super-rich pay their share of taxes: You try to change the law.

Seniority has been the compromise of choice before.

It's hard, but "the law enables corruption so we can do nothing about it" is a defeatist attitude.

To what? How are you going to measure merit?

It depends entirely on the situation.

I work with computers. I know that the compiler doesn't care about your skin color. The linker doesn't give a damn what genitals you have or how long your hair is. Shared objects don't throw ABI compatibility errors because you have a certain religion. Obviously judging merit in computing is a lot easier than in other fields, but it's a start.

Lol, I didn't say "do nothing about it", I already listed two other metrics that teacher unions are using instead of merit.

Scientific research manages to get around such biases with double-blind studies. As long as people making the decisions are aware of traits that are controversial but not germane to their employment status, people will be making biased decisions.

@forest Wasn't there some papers awhile back on how some algorithms (facial recognition and I think mortgage qualification?) that turned out to be racially biased due to poor QA/testing?

Oh, there are lots of those. It's not due to poor QA or testing, but due to the difficulty in finding a truly unbiased dataset.

You have to make sure there are no proxy variables by which the ML can make a decision based on a variable that should be hidden to them. That's why using ML is so dangerous. There are perverse instantiations everywhere.

That's why it's so horrific that the US court system uses ML to determine parole eligibility.

Such systems see the prison system filled with violent black criminals, and when a non-violent black criminal is up for parole, the system concludes that they are too dangerous to be let out. Meanwhile when a violent white person is up for parole, the system grants it. It's not intrinsically racist, but it is reinforcing existing problems (such as biases present in the criminal justice system). It doesn't realize that the prison system begets violent black criminals due to its racism!

The neural network has no idea that putting a person away for longer because they're black when they were guilty of a victimless "crime" such as drug consumption will radicalize them and lead them towards a path of criminality, given that they're now ineligible to work in many industries by virtue of being a felon. The system thus "concludes" that blacks are violent and can't be let outside of prison, which reinforces systemic racism in the prison system.

ML is great, don't get me wrong, but even the best NN is nothing but a fancy curve fitting algorithm.

Who would've thought that isolating criminals from society would encourage them to pursue a living outside of societal mores?

ikr

No one could have predicted this.

And when towns give less funding to black schools and black neighborhoods always get the short end of the stick, people wonder why so many turn to crime. It's a vicious cycle. A naïve observer would logically conclude that blacks are more likely to be violent than whites, which is technically true, but they would also easily and fallaciously conclude that it is due to some innate, intrinsic tendency to be violent, which is racist and untrue.

ML algorithms, no matter how much we try to align them with reality, will come to the same, dangerous conclusion.

They won't stop and think "hm, maybe some external pressure is causing this and it's become a vicious cycle?" No, they'll just decide that the "traits" of blackness are reliable predictions for the traits of criminality. The result? More cycles of violence and fatter wallets for those who invest in for-profit prisons.

I'm planning a visit to California and I'm debating whether I want to try cannabis since it's legal for recreational use over there. But it's still illegal in all cases in Texas, iirc.

Texas can't criminalize your use of it in a state that permits it.

Considering marijuana stays in your system for months, iirc, it's entirely possible. Though it's unlikely that I'd be subject to a random drug test.

It's not illegal to have it in your blood.

I mean, your employer might get upset, but you can't be charged with "having marijuana in your blood" (if you're not impaired).

@Fredy31 I thought the name was Eric

Which is coincidentally also the name of one of this children

U.S. Representative Liz Cheney, a fierce Republican critic of Donald Trump who has played a prominent role in the congressional probe of the Jan. 6 assault on the Capitol, lost to a Trump-backed primary challenger in Wyoming on Tuesday. reuters.com/world/us/…

The GOP and its supporters are truly beyond all hope

> With 99% of expected ballots counted in Wyoming, Hageman led the Republican field with 66.3% of the vote, followed by Cheney with 28.9%, according to Edison Research, an election monitoring firm.

That's a lot of people voting to support a candidate in favour of trying to illegally seize power

@Wipqozn Thing is, I would guess a bunch of those 28% cheney voters will not vote GOP next time around

My copium tells me that as long as the GOP continues like that, they will lose more and more, until they lose basically any power they hold

'Fk off you fkwads': Peterborough Mayor says to QAnon followers who tried to arrest police toronto.citynews.ca/2022/08/16/peterborough-mayor-qanon

Haha fucking chat markup