The meaning is likely understood/inferred by many if not most, sure.

It's still a "contresens" (can't find the right word in English, literally counter to its meaning), and should absolutely be avoided for clarity.

Let's not just say that it's alright

It's alright. Human languages aren't really logically tight the way computer languages are.

An example that goes completely unremarked on is "near miss", which logically means something that came close to missing but actually hit, and yet in idiomatic use means the opposite. People also get upset at "literally" to mean "figuratively", another one I find strange because it's an intensifier.

Clarity matters more in formal writing, and "couldn't care less" isn't particularly formal in any case.

I wouldn't put these in the same category. The inversion of "could care less" meaning "couldn't care less" or "unloose" meaning "loose" are similar.

But "near miss" is more a parsing ambiguity, if not a mere disagreement about grammar. People who think it is illogical seem to assume it is "nearly missing". But in actual usage it is more that "near miss" is like a "narrow miss" and a "far miss" is like a "wide miss", all encoding distance to the implied target/hit zone.

I did use literally correctly.

And I can't agree with you. As a non native speaker, I deeply appreciate people making an effort to use language correctly to transmit information. I call that being mindfull of your interlocutors.

I'm also a non-native (though near-native) speaker and writer. I grew up reading a lot of English but not speaking much of it.

In a way there’s nothing wrong with ”near miss”. It’s a miss not far from the target. Still a miss.

George Carlin had a bit about “near miss” and other illogical phrasings.

It is alright. Most people can figure out from context clues what the writer means and the only thing being pedantic and demanding about other peoples’ language does is make them REALLY not want to do what you’re saying.

Sounds vaguely similar to Jesperson's cycle and double negatives, the "couldn't care less" idioms. And "absolutely avoided for clarity" is a bit harsh, language is by its nature imprecise and telling people how to speak has (thankfully) almost never worked to avert language change.

https://en.wikipedia.org/wiki/Jespersen%27s_cycle

Objective reality, it exists.

Telling your audience obviously false / anti-factual lies, without any regard for fact checking, is not just "biased reporting". And it is inherently wrong, malevolent, evil.

Anyway, I'm amazed each time I hear right wingers who did not get the joke seemingly complaining about how Reality has a left leaning political bias...

Lying is not the same as bias. That is the problem, people often conflate the two.

Yes, permanently. Sonnet 3.7 is already number one in the ranking. Grok3 has no API yet.

Nice !

Could we please get benchmarks for architect / DeepSeek R1 + claude-3-7-20250219 ?

To compare perf and price with Sonnet-3.7-thinking

I don't think that's what the words value and worth mean.

It seems obvious to me that the value of a book teaching me something is not its price, and if that book was free it wouldn't be a worthless book.

insisting on monetary value as the only benchmark of worthiness looks to me as a kind of misguided religion of Capital.

The Open(Closed)AIs of the world have millions of dollars to spend on IP datasets.

Arguing that copyright forbids training AI models without paying authors is the moat that would prevent any hope that small labs, individuals, and open source communities can ever compete with these huge corpos.

The books and other artworks they are arguing over didn't come generated spontaneously from nowhere, and it's disingenuous to refuse sharing them to inform what is basically the worlds next currency : Intelligence. Doing so is just saying that knowledge and intelligence must belong to rich corpos only and never be democratized.

This is unexpected but Meta is basically being the good guy by giving away their research papers and models weight resulting from millions of $ of training.

The alternative to open source AI is everyone's subjugation to the oligarchists in charge of Intelligence. Copyrights holders who argue against free training of open source models from their work are morally and ethically wrong here.

In any case it's really easy to configure a BitTorrent client to never seed. It would be dumb not to when torrenting for work.

Steam can fill up much more

I'm getting my Steam games at 2Gbps, and I am suspecting that my aging ISP's "box" is to blame for the cap (didn't want to pay my ISP for the new box that officially supports 8Gbps symmetrical, and just got a SFP+ adapter for the old one). I pay 39€/M for what is supposed to go "up to" 8Gbps/500Mbps on that old box.

Games from Google Drive mirrors are coming at full speed too. Nice when dling that new Skyrim VR 90GB mod pack refresh

Where I'm from, racial statistics and profiling are forbidden since they're antithetic with republican universalism ; but even taking the US culture into account, this point of view seems extremely bigoted.

Can you explain why you find it bigoted? I think if you ask a lot of Asians you'll find they agree with the statement.

"childhood household income. That transcends race or gender"

vs

"targeting childhood household income will see representation of the groups you're trying to help drop to nearly zero"

Did you misunderstand the point? It seems right that if you want to help poor people, you should target poor people, how could that be counterproductive?

I think GP's point is that if the goal is to help those disadvantaged by childhood poverty, assuming similar levels of poverty, it will help white children who are relatively less disadvantaged and not help relatively more disadvantaged black children.

But that should be the goal. Using skin color as a proxy for being disadvantaged works only inasmuch the proxy is precise enough. If it's not, it's just another bias.

If you had a better metric that can be used to help people from marginalized cummunities, wherever they happen to be, you should be using those if the goal is to be more inclusive, more diverse and more equitable.

But the problem is that the very same psychological mechanisms that drive racism are the ones which drive these modern attempts at fixing racism.

I think there are two different but related angles at play:

1. Historical injustice. Racism against black communities in the USA has such a long and disgusting history that when new generations learn about that it's understandable that people want to wash that away and find ways to make amends and counterbalance things.

2. Since we live in a world where skin color is a visible marker that you literally wear on your skin, there is a sense that you're just being disadvantaged for being you and thus you need a more-than-normal counterbalance to set things straight.

We thus ended up in a situation where we destroyed the aspirations of a truly color-blind society. We need to keep reminding ourselves about the fact that color matters.

This is not helping defusing racism. This is feeding racism because despite the best intentions it operates in the worldview of racism.

Of all the words you Americans have purged from the language you kept the word race, a word that post WWII many European nations successfully defused.

You seem to be reasoning from the standpoint that anti-black racism is no longer a factor today.

This is not the case.

Once you understand this, it is easy to see that actual, first-order racism is what destroys the dream of a colorblind society.

I understand that anti-black racism is still a factor today.

I really fail to see how doubling in on the importance and reality of race and skin color in particular (as opposed to culture) everywhere is going to make that problem go away.

Exposure to other races through integration increases understanding and reduces the tendency towards prejudice.

Makes sense?

Skin color is visible, but it is not a marker unless you allow it to be one.

But what if the poor people have the wrong skin color?

...raise taxes on the households of whatever the "correct skin color" is until they qualify for these programs? Obvs. :)

/s, I hope. :/

I don’t know. When Sweden reached 58% women in university, they stopped the gender equality programs. Why would you help men if everyone knows they are advantaged?

(Oh man, please answer, but we’re overdue for an overcorrection, and it might look as dangerous as 1934).