curious_straight_ca
No bio...
User ID: 1845
curious_straight_ca
1yr ago
Only like 1 in 20 of the frequent posters here are actually white nationalists
curious_straight_ca
1yr ago
Most people don't feel like writing long toplevel posts most of the time, and the BLR isn't allowed because at least 'long effort post' is a content-neutral filter and the mods don't want to have a filter that isn't content-neutral (both for philosophical reasons and because it'd take more mod effort). I think this is a mistake.
Also, is there so much interesting stuff going on right now? Seems about usual to me.
Our immigrant pool is ... fine, certainly not awesome. It could be so much better than it already is if US immigration was intentionally administered in the interest of good immigrants. IQ tests, demonstrations of technical skill, unlimited in number but very expensive paid sponsorships, maybe with a culture exam or something if you care about that. Instead, there's generic administrative stasis and a political tug of war between 'poor mexican immigrants :(' and 'And Some, I Assume, Are Good People', and only minor improvements get done by pro skilled immigration interest groups.
curious_straight_ca
1yr ago
Feels more like it's earnest but bad output as opposed to chatgpt, it just doesn't feel like chatgpt's cadence or grammar? But maybe it's prompted with that style or something idk. Not sure. (edit: either way, good decision)
curious_straight_ca
1yr ago
Yeah. Evolution really really wants you to have kids (and you probably should!), so you'll have a strong drive for a romantic relationship independent from anything else.
curious_straight_ca
1yr ago
From what I vaguely know and from skimming the wikipedia article, I'd guess that it was obvious to men on the ground, but not certain. So some fighting did happen because they wanted to push forward in case the war restarted, but it was obvious that wasn't the most likely outcome. Even granting that point, one can still blame the leaders, who had the option to pause earlier but chose not to. And it's not just my position that this wasn't necessary:
After the war, there was a deep shame that so many soldiers died on the final day of the war, especially in the hours after the treaty had been signed but had not yet taken effect. In the United States, Congress opened an investigation to find out why and if blame should be placed on the leaders of the American Expeditionary Forces, including John Pershing.[41] In France, many graves of French soldiers who died on 11 November were backdated to 10 November.[38]
curious_straight_ca
1yr ago
I think the point is that it was an obvious prelude to peace, and further fighting forseeably ended up accomplishing no military objectives whatsoever, which strongly implies pointless deaths that could've been easily averted.
I wonder what the correlation between selfreported IQ and orientation is in scott's survey.
edit: 2022 seemed suggestive for gays having higher iq but no consistent pattern over all of the past surveys. there's something suggestive now for gays having higher sat verbal but lower sat math than straights but that'd probably disappear with a larger sample
If you're going to make a toplevel post about "are gays smarter", IMO you should at least type the thing into google scholar.
This paper (which I haven't even skimmed) claims homosexuals are on average more intelligent. I could see it being true that smart people are more likely to apply that intelligence to their actions and thus be 'weird' and either actually be homosexual, or to engage in homosexual-ish behavior without having the usual innate cause (if that exists). But, eh, one paper of the usual quality isn't enough to believe anything this fuzzy.
... I wish there was just a 'biobank' dataset I could download and query for the correlation between IQ reported sexual orientation. Publication bias, p hacking, would instantly disappear as concerns. I'd probably vote for a law that nonconsensually released all of my demographic, medical, and other useful data along with everyone else's (it wouldn't be quite as useful if it was opt-in, but it'd still be useful).
I disagree with enough of the content I've read that I've fully separated my appreciation of a joke/piece of writing/etc's technical merits from my appreciation of its accuracy. Also, it's quite fun, and very informative, to try and see if you can inhabit the mindset that produced stuff that's that wrong. This also (but less so) extends to other writing styles.
curious_straight_ca
1yr ago
Maybe Mine Were of Trouble: A Nationalist Account of the Spanish Civil War by Peter Kemp? I haven't read it, but it seems to be more of a personal narrative about Kemp, foreigner who joins the Nationalist Army and becomes an officer, than it is a tome of history. It's been a big thing on dissident right twitter for a while, to the point that the edition I linked is a reprint by "Mystery Grove Publishing", a "far right publishing house". "Everyone on right-wing twitter loves it" is, if anything, negative evidence about the quality, but it seems to have good reviews from other normal people who wanted to see the other perspective.
https://samkriss.substack.com/p/live-from-the-hate-march
It was Armistice Day last weekend. One hundred and five years ago, on the 11th hour of the 11th day of the 11th month, the First World War ended. They’re very pretty, all those upright ones. A good piece of trivia for schoolchildren to learn. But the armistice was actually signed at 5 am that morning, and to get that nice symmetrical figure the war had to keep going for six more pointless hours. The Germans had asked for hostilities to end immediately; they were refused. So shortly afterwards, an American unit tried to cross the Meuse under heavy fire. British artillery corps, reasoning that it would be more expensive to lug their shells home than to simply fire them all now, spent those last six hours unleashing one last hellish barrage up and down the German lines. French soldiers stormed occupied villages, and many of them died. Nearly three thousand men died on the 11th of November, considerably more than on any average day on the Western Front. Most of their graves say that they fell on the 10th instead. Better that than the indignity of dying for no reason, for no objective, for nobody’s advantage, to change nothing, in a war that was already over but still kept on churning, kept chewing through its victims, simply because war is what there was, and it needs no other justification than that. Those six hours were war in its purest form: no politics, just total sourceless hostility. The moment was coming when the people in the other trenches would simply get up and walk home, and to fire a shot at them would be murder—but right up until that moment, you could still get your sport or pleasure from pinching out a stranger’s life. Perform the rites of the White God. There were generals who wanted their last chance at glory; soldiers too. The last man to die was one Private Henry Gunther, an American. A few months previously, he’d been knocked down from sergeant for complaining about conditions on the front; he wanted to get his pips back, and he didn’t have long. Alone, he charged a German machine-gun nest with his bayonet. At first, the gunners tried to shoo him away. But he kept running at them, so in the very last minute of the war they shot him dead. Afterwards, his rank was posthumously restored and his corpse was issued a citation for gallantry in action, along with the Distinguished Service Cross. Maybe the worst thing about war is that sometimes, you really can get exactly what you want.
Mostly it's just great writing, but it doesn't belong in the main thread because I have nothing to add (edit: also, now that i've actually read the rest, it doesn't belong in the main thread anyway, and I don't agree at all, but it's insanely good writing). It almost seemed too grimly poetic to actually be mostly true - that was, initially, going to be my Small Scale Question, or maybe "is it misleading somehow" - but it checks out to like 2 minutes of research.
So, uh, genuine small scale question: If someone wants to post something here that doesn't quite fit into one of the themed threads, but where they can't muster a full toplevel post, where does/should it go? Are we missing a whole class of posts like that?
(The rest of the article is more directly CW about palestine protests, and also quickly devolves into lurid hallucinations)
curious_straight_ca
1yr ago
I don't think it does, given it's substantially stricter than every other platform, including Youtube, Tiktok, pre-musk Twitter, and similar. I think it's a combination of - economic incentives mean management has some pressure to censor, and then the people they give the 'trust and safety' or w/e authority to are very progressive.
curious_straight_ca
1yr ago
Reminder to anyone reading this: If any of these books sound vaguely interesting, you could be reading one of them in approximately 90 seconds! Go to https://annas-archive.org, type in the name (and author if the name isn't specific), click the right one, click one of the links to pdfs or epubs under 'slow external downloads', and then get to reading! Don't feel bad about piracy, you weren't going to read it otherwise, and some of the authors are already dead. You can always buy a real copy later.
Actually, you may need an epub reader application on windows, I don't think it has one built in.
curious_straight_ca
1yr ago
It's been remarked on here before, but it seems nobody particularly cares about FtM trans people. They blend in easier and - more importantly - don't seem to threaten anybody either in physicality or status.
This used to be true, but in the past few years "our female children are being abused by teachers/the medical system/tiktok and transitioning and mutilating our breasts" has become a huge thing.
Most of the locals here are what most progressives would describe as transphobic! "Physician, heal thyself."
Personally disagreeing with an idea doesn't really make me worse at arguing for it, though. I think the most effective 'cure' is having social exposure to trans people who are reasonable and friendly. Other than that, recommend highlighting stories of trans kids being kicked out of their homes by their parents when they were outed and the emotional pain of individuals' personal struggles with gender dysphoria.
edit: Oh, I read some of the below replies. Frankly, trans people aren't any more viscerally disgusting than fat people, and you'll be exposed to a lot more of those in most parts of America. Like anything one finds disgusting (justified or not), you'll rapidly learn to ignore it after being exposed a few times (and that doesn't have anything to do with your political opposition to trans), so I wouldn't worry at all about it. Also, unless you spend a lot of time in the rationalist community, you probably won't run into enough trans people to be worth noting.
curious_straight_ca
1yr ago
Probably not, given people don't really like her.
Leaving aside the question about whether a VP has ever been more popular than their president2, it’s remarkable how little enthusiasm there is about the prospect of a Harris candidacy or presidency. I’m not particularly well-connected among Democratic Party establishment types — it’s just not my crowd. But between the conversations I have had with people in those circles and my “normie D” friends, I don’t think I ever heard a single person advocate that Biden should settle for one term and let Harris run instead. Hell, even in her column that did advocate for Biden to step down, New York Times columnist Michelle Goldberg — who I find to be a consistently thoughtful reader of progressive sentiment — conspicuously mentioned that the alternative didn’t have to be Harris, citing Harris’s poor polling.
... It’s also a pretty consistent pattern; Harris polled worse than Biden in literally every poll that I could find. And I’d note that it’s not just a matter of Harris performing worse against Trump because she’s less well-known and so more voters flow into the undecided column. Instead, Trump actually gained vote share against her, getting 46 percent of the vote against Harris as compared to 44 percent against Biden.
... However, Harris has run for president before and it didn’t go well — it didn’t go well at all. Considered one of the frontrunners5 for the Democratic nomination when she launched her campaign in January 20196, Harris wound up dropping out of the race in December, well before the Iowa caucuses. It was in the Scott Walker/Jeb Bush/Phil Gramm/Ed Muskie/Rick Perry tier of epic primary season flameouts.
curious_straight_ca
1yr ago
If the entire thing was a troll to see how much effort we'd spend engaging with a silly but elaborate hypothetical, it'd have been a huge success.
Also a big point towards having a BLR or more toplevel posts or something, because there's clearly a lot more demand for stuff to reply to than supply of stuff to reply to.
All of the big LLMs are based on transformers. You said "Ultimately what a regression-based machine learning algorithm (of which LLMs are a subset)".
curious_straight_ca
1yr ago
The obvious "nudge" style suggestion is to put on closed captions / subtitles for every youtube video / TV show they watch. Or get them into video games or websites that involve reading a lot of text?
curious_straight_ca
1yr ago
While this sort of statistical processes can excel at associative tasks where the bounds of likely inputs and outputs are known in advance such as linguistic translation and ranking search results, it ends up being worse than useless for other more agentic tasks like pathfinding, and is only capable of "finding useful information" in so far as what is "useful" and what is "statistically probable" based on its training data are in alignment.
@HlynkaCG Actually, the techniques used in language modeling are great at "pathfinding" and other "agentic" tasks, too. See Decision Transformers and similar work. One of the most central, and at-the-time most surprising to many, results of ML is that the same techniques work for a wide variety of tasks. Neural nets "want to work".
You say NNs / language models are regression based. This is vacuously true. Wiki says:
In statistical modeling, regression analysis is a set of statistical processes for estimating the relationships between a dependent variable (often called the 'outcome' or 'response' variable, or a 'label' in machine learning parlance) and one or more independent variables (often called 'predictors', 'covariates', 'explanatory variables' or 'features').
So, language models are regression based in the sense that they predict things based on other things. Every possible technique for doing what language models do, or indeed any method of machine learning or AI, or indeed humans behavior itself, could be cast as a "regression", in that sense. In the sense you mean, though, of randomness or simplicity, they aren't - the models that are trained are horrendously complex, and capable of representing very complicated computations. As opposed to "regressions" in the colloquial sense, which are relatively simple statistical models.
What's happening mechanically when you "train" a regression engine is that you are populating that table and assigning different statistical weights to the various outputs within it based on the prompt provided
You're, presumably, familiar with physics and causality, right? Any discrete theory of physics (and modern physics is strongly suspected to be discrete, as involving real numbers anywhere leads to all sorts of paradoxes) can, necessarily, be modeled as a (very large) "table", or matrix, with a row/column for each world-state, and various transition "statistical weights" / probabilities from each state to each state. This is certainly an incredibly coarse-grained representation, especially given you need a state for each large-scale quantum state (distribution-across-universe-branches), but it's doable. So, given your "regression engine" can, in theory, run the entire universe, I think it's premature to say it can't run an AI.
Now, obviously there are chinese room-level scale issues with the comparison, and physics has mathematical patterns that lead to a description much simpler, and smaller, than a transition matrix of size 2^2^(number of atoms in the universe). Fortunately, neural networks have those too! They're not huge transition matrices either, but very complicated functions with a lot of internal regularity.
So, HlynkaCG, I don't get why we keep having these discussions, you just assert a bunch of things that are patently false, and then repeat them a few months later after they're corrected.
curious_straight_ca
1yr ago
Obviously @Amadan's ban for nantafiria was deserved, but imo your tone here was unnecessary and baited him into it. He made a sourced and reasonable post about how the right in europe is, sometimes, able to reduce immigration. You responded with - yes, a piece of evidence - but then a vague bundle of grievances
true. The democracy of western countries is about is authentic as American wrestling, and most of the implemented policies are not even debated with the public. I'd consider that par for the course, but European neolibs screaming their head off about the rise of racism, islamophobia, populism, and "paranoid rightists", while simultaneously bragging about how the paranoid rightist policies are dutifully being implemented, is a bit much.
Later, you accused him of holding a strong position he doesn't hold, on the grounds of 'everyone I dislike has the same opinions', in this case open borders. I can see why he'd be upset.
Ok, so your position actually is "you will get mass immigration, no matter what, and that's a good thing". Just say it with your chest.
curious_straight_ca
1yr ago
Other reports that it isn't true that this contributed to his exit - https://twitter.com/alexeheath/status/1727472179283919032
curious_straight_ca
1yr ago
As to why the same isn't true of e.g. communists? Honestly I have no clue, but I think that indicates a problem with the communists.
I mean, in part because we're much closer, politically, to communists than we are nazis. The idea of uplifting the poor, equality for all, the will of the people, etc, are things that modern americans and communists share. Whereas the nobility of war, a duty to serve the state and its leader, the superior races pushing out the inferior, jews being bad, etc, aren't something we share. Communism is noble but misguided in practice, naziism is evil.
This means communism 'goes down easier' than fascism, so people have a much stronger visceral reaction to fascism. The fascist then has to compensate for that, e.g. by rejecting violence
I don't think 'fundamental constraints of a medium' are real, necessarily, every mottepost could be a TikTok video if there weren't any other options, just like 2000 years ago knowledge was mostly transmitted via spoken lecture or conversation. But as people figure out social media platforms, they'll tend to collect their serious, subtle posts in long-form text, and their idle chatter and jokes on places like Twitter. So even though Twitter has most of the same people people you see posting here, or in the ACX comments, it has a different 'culture', and it's one that's a lot less careful about accuracy, and more just people slinging vibes at each other.
Don't get me wrong, I spend a lot of time on twitter, and have for years. I think a lot of other people here do too. But there's a reason I still come here.
More options
Context Copy link