site banner
Advanced search parameters (with examples): "author:quadnarca", "domain:reddit.com", "over18:true"

Showing 25 of 168 results for

domain:kvetch.substack.com

This only holds for climate scientists trying to come up with new global models. Useful climate science looks like trying to make specific predictions about specific areas on a specific time scale in the context of an extant model, so that human infrastructure can anticipate and adapt to disruptions to established patterns.

(It's the difference between "AI risk researchers" who come up with yet more convoluted thought experiments on how to do timeless bargaining with omniscient gods, and "AI risk researchers" who are actually creating code to interpret and control what's going on inside neural networks. I can see why someone would be fed up with the former, but the latter is actual expert work that needs doing, and - so long as they sub-optimally remain a package deal - justifies the existence of the overall field.)

Why don't you think that climate science is useful science?

Many kinds of scientists are irrationally prone to thinking their particular specialty is the most important thing evah, but AI risk is still a real thing, space exploration is still a real thing, deadly viruses are still a real thing, etc. I think the same is obviously true of the climate. Global warming isn't going to literally set us all on fire by 2035 but climate change is still an ongoing phenomenon with massive global implications and it needs to be studied. So long as you don't follow them on X, I think most scientists still produce more light than heat, if you'll forgive the expression.

And if the Earth gets warmer this isn't a crisis

Come on. It would suck for far more than "penguins and polar bears". It'll suck for the current balance of worldwide agriculture, it'll suck for coastal population centers, it'll suck for vast swathes of land that are already near the threshold of unlivable, and, oh yes, it'll suck for all the currently temperate areas which will inherit the latter's status as the arid "well, you can eke out a living there, I guess" hell-holes. This is true even if you're correct about global warming opening as much hitherto-frozen land for settlement up north as it will ruin further south. We'd be looking at a major reshuffle in what countries control what kind of territory and resources, which may not favor the West much. (Indeed, odds are it wouldn't: we'd been dealt a good hand already, to the point that some view sheer luck of the draw on Europe's local climate as, if not the secret to Europeans' worldwide success, then at least a crucial prerequisite. We can really only go downhill.)

Even if it turns out net positive in the end, it needs to be anticipated and planned for in order to mitigate the damaging side-effects of the disruption itself. Which is exactly what I had in mind when I spoke of "global implications" and why we need climate scientists. I agree that the Earth getting warmer isn't x-risk. But there's far more to whether a crisis is worth averting or mitigating than whether it'll literally wipe out humanity. (Particularly if we're talking about a specific country's incentives rather than Homo Sapiens's as a whole.)

Well, they've gotten better and better over time. I've been using LLMs before they were cool, and we've probably seen between 1-2 OOM reduction in hallucination rates. The bigger they get, the lower the rate. It's not like humans are immune to mistakes, misremembering, or even plain making shit up.

In fact, some recent studies (on now outdated models like Claude 3.6) found zero hallucinations at all in tasks like medical transcription and summarization.

It's a solvable problem, be it through human oversight or the use of other parallel models to check results.

Low performers are irrelevant, it's high performers that are dangerous. Who is more dangerous as a grudgebearer - Joshua VerbalIQbaum or Mgubu the Witless? Likewise it's not unreasonable for Chang, Zheng and much of the Maths Olympiad phenotype to hold a grudge for their treatment in the 19th and early 20th century. You can always ignore Mgubu, he has no armoured brigades or advanced rhetoric.

There’s nothing wrong with saying that you find a comparison ludicrous, but we ask that you leverage a more substantial complaint than “TDS.”

The category "African-American" is neither limited to American Descendants of Slavery nor does it include Elon Musk. The Census definition is "A person having origins in any of the Black [sic] racial groups of Africa."

Not only is this silly in that it declares government to be the final arbiter and definer of race, it doesn't even work out well when governments themselves disagree on these types of definitions.

Check out the status of groups like the Brazilian or Portuguese Americans for instance.

For the Portuguese, whether or not they are officially Hispanic depends in part on the state they live in.

Some states, such as Florida, categorize Portuguese Americans as Hispanic, while others, such as California, do not. In a few places, including Massachusetts, laws and regulations treat them as a disadvantaged group for at least some purposes.

Is ethnicity really a concept that is state defined? Does a person's ethnicity really change if they move from California to Florida?

Brazilians are even more confusing, they aren't officially considered Hispanic or Latino either, and yet it seems more than two-thirds of them define their identity that way

This is more people than some traditional Hispanic groups!

In fact, enough Brazilians identified as Latino in 2020 that they would fall in the middle of rankings of U.S. Hispanic or Latino origin groups by size, if they were officially counted as one. In 2020, Brazil would have been the 14th-largest Latino origin group with 416,000 who identified as Latino, ahead of Nicaragua (395,000) and below Venezuela (619,000).

It even changes constantly in this list of racial prerequisite cases on who is determined white

Syrians, Asian Indians and Arabians are both white and not-white. Mexicans btw in the single case for them are white.

Chinese aren't white, but they do get a pass in Jim Crow Era Mississippi to attend the white only schools and join the White Citizens Council so it seems even that isn't fully clear!

The alternative is exceedingly simple. "African-American" is commonly understood to mean American Descendants of Slavery, not Elon Musk.

We can not expect every single person to have been exposed to everything and culturally absorbed this in their life, especially when it is so unintuitive. Like that XKCD comic that's pretty famous, there are a lot of people out there who don't have or never experienced "everyone knows" situations.

So if your category is unclear and inconsistent, and other more intuitive interpretations clash with the culturally accepted one, then you are inevitably going to land on a bunch of people who use it "wrong".

You're just "African", and that's how it will be until race/ethnicity-based affirmative action schemes are totally abolished.

So he's an African, and he's an American, but he's not an African American? Certainly you can understand how that at face value looks incredibly stupid right?

Have not doesn't mean they will not.

Yes, but do a sufficiently bad enough job at the bridge that they tell you to stop. Sometimes the only way out of an impossible situation is to act so incompetent that people stop forcing you to do it (after all, if you hadn’t saved the guy the first time, you wouldn’t be asked back).

Couldn't your conclusion that 'If Hamas manages to get an attack off it's the entire host nations problem as well' apply to Iran giving them a nuke in the first place?

No, because Iran is the only one capable of retaliating (in a nuclear fashion) hard enough to discourage that. And Israel doesn't need to go nuclear if this happens; a conventional war would be just as destructive for these nations and peoples. Perhaps that is part of why the neighboring countries are unwilling to host the Hamasi as refugees.

Couldn't Israel just state preemptively they will regard any use of nukes by hamas as use by Iran

Maybe, but I don't think Israel can win a war with Iran (hence the emphasis on keeping them down/contained). They're sufficiently equipped to wreck any country Iran allies with outside of that lovely mountain range that defines the western Iranian border, but unless the Americans want to put their boots on the ground and suffer the 3:1 attacker casualty rates to conquer Iran then Israel can't really touch them. Israel doesn't have those numbers, Iran's a peer nation (except for the nuclear weapons), and if either tries in the future are the Israel-hating Blues (and even Reds; Israeli influence might not be as stable in an era of Red reforms) even going to lift a finger?

Remember, the ultimate problem Israel is fighting is that, absent Rome/Europe/Washington and its religious fixation on holding Jerusalem, it is the natural geopolitical state of Judea to be in the Persian orbit. Hence the rhyme with Biblical times- Jews evict the Canaanites, then the Persians conquer the Jews.

Which climate scientists have made a prediction including a date of extinction for the human race, particularly one with a date that currently counts as definitely falsified (presumably in the sense of having already passed)?

Good, didn't know about this.

Great way to promote ambiguity, though only very naive people will fall for it.

That's what I am saying.

But the only way to do it is thermonuclear weapons.

20*20 kilotons on Israel would be catastrophic but it'd be survivable for Israel.

It'd not be survivable for Iranian government.

100*400 kilotons would destroy Israel, possibly even partially prevent retaliation.

That's the moment when Iran would have true deterrence and MAD with Israel.

Being able to wound the enemy and then assuredly die is not deterrence.

Out of boredom, I'm using Gemini to make a mortar calculator app that takes in screenshots/grid coordinates and outputs firing solutions. Should work in theory!

I expect that when people usually say that, they're implicitly stating strong belief that the problems are both solvable and being solved. Not that this necessarily means that such claims are true..

Problems with CO2 in the Southern Ocean? Need it quickly absorbed? One man has the most based idea of all time: https://arxiv.org/html/2501.06623v1

True. But really, being proud that you reached agriculture and tribal-level development isn't very impressive. Only a few thousand years behind the curve on metalworking! One wonders whether formerly-Aztec Mexicans or Mayans are snooty about being lumped in with mere nomadic 'native Americans' who never got that into astronomy of stone-working.

I am extremely skeptical at that claim. I mean, surely, if you examine LLM at what humans are usually examined at, things that are hard for humans - like perfectly recalling bits out of huge arrays of information - it would probably do pretty good. However, at things that human are never examined it - like common sense - because most humans that got through law school would have it, otherwise they'd fail out and probably be either institutionalized somehow or ejected from the society in some other way - LLMs are still terrible.

Just days ago I tried to use LLM advice to configure a scanner on my Mac. It managed to give me ton of advice that didn't work (because it kept hallucinating and confusing different Mac models) but then it managed to give an advice that seemed to work. I stupidly followed it. It broke my Mac completely. I decided to take hair of the dog approach and asked the same GPT for the fix advice. After another hour or so of hallucinating and meandering, it managed to make the problem worse. Then it had me to try a dozen or so non-working solution, each one ending with congratulating me on discovering yet another thing that doesn't work on my Mac - this despite me telling it upfront which Mac it is and it being aware to quote the exact source that says this wouldn't work - but only after suggesting to me repeatedly it would 100% work for sure. Eventually, it started suggesting to me deleting disk partitions and reinstalling the whole OS - while claiming this can't hurt my data in any way, everything would be OK - and I decided to call it quits. I tried to fix it using my wits alone and plain old internet search, and was able to do it in about 15 minutes.

This was a low risk activity - I actually had pretty recent backups and all important shit I have backed up in several places locally and online, so if it killed my Mac I maybe would lose some unimportant files and some time to re-configure the system, but it wouldn't be a catastrophe for me. Now imagine something like millions of dollars, or decades in jail, or the entire future of a person is on the line. Would I trust a machine that claims X exists and solves my problem only to cheerfully admit X never existed and even if it did, it couldn't solve my problem a minute later? Or would I trust a human that at least understands why such kind of behavior is unacceptable, in fact, that understands anything and isn't just a huge can of chopped up information fragments and a procedure of retrieving some of them that look like what I want to hear?

Sorry, I can't believe this "as good as a fresh graduate" thing. Maybe I can believe it's "as good as a fresh graduate on things that we check on fresh graduates because they are hard for fresh graduates so we want to make sure they are good" but that misses the obvious pitfall that things that are very easy for a fresh graduate - or any human - are very hard for it in turn.

people of slanted eyes are too.

Are smart enough? I'm parsing your sentence but the general tone seems dismissive, whereas this seems complementary. The slant eyes bit is an odd moment of bluntness for you.

Hamas is primarily an enemy of Israel

This is not an uncommon position, but it is clearly incorrect. At the very least, someone who makes this line of argumentation needs to give a disclaimer to avoid being correctly called a liar. That disclaimer would be something along the lines of, "because of Hamas' strategic and military incompetence, and the vast distance between us and them I don't consider them a threat."

I don't find these arguments FOR incompetence compelling, but if you are adopting them you should be clear about it. Because I know in my heart that if Hamas had our army and we had Hamas's militias, they'd simply kill us all with nukes and laugh while doing it.

I did not find French 'easier' by knowing Latin and Spanish. I suppose they made it easier to learn each other, but French is pretty different from the other Romance languages grammatically and in pronunciation.

Thanks for the info!

Very interesting stuff.

No that is not the lesson, there is nothing to fear from a "low performer". What you need to fear is the person or group who you dismissed as low performing but have the potential to not be, because if you fuck em there is a good chance they'll fuck you back and you will deserve it.