@username's banner p

username


				

				

				
0 followers   follows 0 users  
joined 2022 October 03 19:47:08 UTC

				

User ID: 1468

username


				
				
				

				
0 followers   follows 0 users   joined 2022 October 03 19:47:08 UTC

					

No bio...


					

User ID: 1468

I'm probably misunderstanding georgism, but would it even help without YIMBY/zoning deregulation? If land value prices in zoning restrictions, then a plot of land that can only be used for single family housing won't be worth more than a single family can afford. If land value is independent of zoning restrictions, then it just makes a ton of people poor without letting them build improvements that house more people.

I've come across HBD a few times in these CW threads since themotte moved off reddit, and dozens of times before that. As far as a I know this is basically the only place you can openly discuss it.

This HN discussion is similar to what you'd read here, but shifted to the left. Here HBD is basically supposed "true" and those who disagree are in the minority. We don't argue particularly better or worse than the HN comments. Often, it's just some top-level complaint about the blank-slate view and others who agree.

I'm not a fan of the vaccine mandates but there's a couple of things I think we should keep in mind here. These facts combined don't justify the mandates, but it does explain some of the situation.

  • Vaccine efficacy is the relative risk difference of infection, severe illness or death (3 different measures) between a vaccinated and control groups, over a set duration. This is the standard way to measure a vaccine's efficacy and it doesn't take transmission into account. It obviously doesn't take into account what happens after your experiment duration is over. So we quickly found ourselves estimating efficacy by looking at hospital admissions and vaccination base-rates once experiments were finished.

  • The vaccine seemed much more promising against the initial strains of the virus. If I recall correctly it prevented ~95% of infections for a few months. Such a strong efficacy against infection does a lot to prevent transmission.

  • There is reason to believe a milder case results in less transmission. You're spreading for shorter periods and expelling smaller viral loads. There was evidence of this. (Admittedly I don't think this is significant enough)

  • Testing for transmission reduction would have been infeasible. There is no standard objective way to measure this. Even with infinite money and without red tape it's not clear whether we should have counted covid particles in the infected's breathing air or something like that, and we couldn't have confidently turned into a "X % transmission reduction"

  • Regarding vaccinating during a pandemic, maybe Bossche is right, we'll never know. However, so far, it looks like the escaping variants we got mostly came from areas that weren't vaccinated. Perhaps it would be less risky if we didn't "meddle with nature" but we were rightly confident that the vaccines would save many vulnerable people's lives.

That doesn't sound like a crazy position to me. I think the lockdowns are a proximate cause, e.g. the inflationary policies of covid-aid funds weren't necessary without lockdowns. The tech-sector wouldn't have been able to become over-valued so quickly if it weren't for everyone staying home.

On the other hand, we don't have a lot of good data for what would have happened if we hadn't done lockdowns. Places who did fewer lockdowns still have to deal with the overvaluation of tech, global supply chain issues and cost of housing increases. So we can't really blame these issues on the marginal lockdown. (Although I'd be curious to see how if places that didn't lock down have better flu/RSV situations)

I'm not even sure who we can blame for lockdowns. Places like Sweden who avoided mandated lockdowns still saw large segments of their economies "shut down". Are ordinary people to blame? There's also very little variance in the responses from different countries/institutions, which suggests 'elite culture' bears responsibility.

And then of course, there's the virus itself. It's easy to say the world could have reacted better, but it's hard to imagine we could brush off covid as a bad flu season. It's difficult to avoid both a large number of deaths and borrowing from the future.

If factory farming is unnatural, is 19th century farming also unnatural? Where do you draw the line? Seems to me you could argue all food, being a product of agriculture, is unnatural, which kind of makes the label pointless.

At the very least, lab-grown meat is a big deviation from the status-quo. I'm not sure it would even count as raw/unprocessed food.

Lots of interesting comments are pointing out that CP, including potentially lollicon, can sway the "marginal" pedophile. Do we have any evidence true marginal pedophiles exist? I'm not talking about those who have urges but don't act because of incentives, but sleeper-agent types who would develop urges if they sufficiently engaged with the idea or pornographic content.

My impression had always been that true paedophilia was some kind of mental illness that was much more binary and that "marginal pedophiles" were rare. People who are attracted to teenagers (Ephebophiles?) would fall into a different camp, which would be a lot less binary and have many marginal members.

I'm not basing this model on any real knowledge, but if seems to me the anecdotes of people being molested are usually about younger children, not teenagers. If marginal pedophiles were a major concern, I'd expect higher frequencies of molestation in older children/teenagers.

It seems to me politics are especially bad at countering inflation. People may be pissed off the "government isn't doing enough" to fix it. However, the levers the government can easily pull to fight inflation (raising taxes/interest rates, cut transfers/subsidies, lower consumption/wages) won't gain them any popularity points. Even worse, there are plenty of socially-desirable policies that'll make inflation worse: tax cuts, raises, loan forgiveness, more transfers. I'm grateful our system is even capable of doing unpopular things like raising interest rates. If monetary policy had heavier democratic influence, I'd worry high inflation would just be the long-term equilibrium.

Ontario's leader is under heavy fire for playing hardball and preventing educational workers from getting large raises. I don't know the details well, but couldn't one argue this is what fighting inflation looks like?

If by liberal you mean left, then the argument is essentially that people are biased against those looked over minorities, (e.g everything from overt racism to implicit bias). How do we know this bias actually exists? The best argument I've seen is Scott Alexander's Social justice for the highly demanding of rigor. It's a little bit outdated. For instance, more recent analyses of the blind orchestra data show it can also be used to support the hypothesis that blind orchestras lowered women's chances of getting in.

They probably want the ability to hit targets with higher confidence. Shooting multiple missiles is an easy way to do that.

Still, I wonder why China ramped up nukes more recently, decades after the soviets and USA have abandoned the same strategy. Also, why is it such a secret? Of course you'd want to keep the details secret, but wouldn't you at least want your enemies to know you have a large nuclear arsenal?

I believe racial groups have different mean IQs and that some of these differences could be partially explained by genetics. I guess that puts me in the HBD camp.

If HBD weren't real, I don't think I'd expect any major differences between countries. Asia and Africa would still be held back by poor institutions. The fact that the middle-IQ group dominated both the lower and higher IQ groups leads me to believe group IQ differences didn't have a high first-order impact on history. I think the biggest differences would be within countries. I'd expect to see more black and fewer jewish scientists, engineers, CEOs, etc. Racism would still exist on a similar scale. We'd worry less about economic disparities, but still worry about representational disparities. American "guilt" towards blacks would more closely resemble european guilt towards jews.

There's another aspect of HBD which proposes that, although men and women have the same mean IQs, men have higher variance than women. Whether that's true or not, I think the counterfactual would be of higher consequence.

I worked in a central government agency and sometimes dabbled in AI policy. When race is blinded, sometimes the computer will still disproportionately target certain races because of correlated factors.

As long as the algorithm is detecting actual errors, it's okay for it to disproportionately impact certain races (although it will still make the news and people will complain). What's more controversial is when the computer is copying existing human behaviour. If human behaviour can be truly racist (e.g. in an irrational way), the computer can inherit that irrationality, even if blinded.

Are you suggesting HBD skepticism would be less prevalent if people understood basic sciences more? I don't think that's the case. HBD skepticism has more to do with social factors than people honestly considering the evidence and coming to the wrong conclusions. The same thing is true for outdoor mask mandates, the reasoning driven by politics and fear (and to be fair, your water-coloring analogy doesn't apply to droplets, only airborne transmission, presumably an outdoor mask would prevent sneezing on others)

This plays into Caplan's argument, in which he admits his economic students go on to support minimum wage hikes, rent control, etc. Presumably these students have intuitive understanding of supply and demand.

All that said, I do agree with your premise. My case against Caplan is as follows: Education is where we go to train the models in our heads. We may forget the inputs, but the models remain. I may not remember all the dates and numbers from the world wars, but learning about those events has improved my models of reality and social structures. That sounds like intuition to me.

Spending plans have been too long for anyone to read for a long time, it's kind of how they're designed.

I'm unfamiliar with the USA government, but in Canada, "spending bills" are all massive and they take weeks/months to put together. The vote is more of a rarely exercised opportunity to veto rather than actually propose changes. Although no one should technically see the full bill before it gets presented to congress, the management board of the government coordinate it with the central politicians for a long time before that.

The truss case is a bit unusual: The inflation hit very quickly and her policies were a very apparent cause.

The average case is a lot muddier. Prices are sticky and it takes a while for inflation to get noticed. Most nations saw a steady inflation increase since 2020. Many of those nations, like the USA, have new leaders since the inflation started. It's probably the case that some leaders picked known inflationary policies before elections to get more votes (e.g. student forgiveness before the mid-terms), knowing the resulting inflation would be delayed and its cause nebulous.

So I finally installed tiktok. While registering, I indicated I was male. I was immediately shown what I can only describe as "anti-feminist" videos, women winning arguments against feminists, jordan peterson interview clips, etc. I generally scroll past these videos quickly, but they got more and more frequent, I probably made it worse for liking a few bill-burr clips early on, but it certainly started very early on.

My wife is a frequent tiktok user, she likes videos you'd expect of women, crafting stuff, recipes, etc. She gets also gets ton of overtly political feminist videos. Neither of us have strong feelings towards feminism. If anything, she's to my right on the gender issues.

I hear a lot of anti-tiktok rhetoric along the lines that china is invading our privacy. I'm much more concerned about tiktok dividing the younger generations and pitting groups against each other. This is probably more algorithmic than intentional, but this effect is almost certainly worse than the privacy concerns. I know this isn't anything new, other social media apps have similar effects, but I think the effect is much stronger with tiktok. With facebook, you inherit the political environment of your friends. With reddit and twitter you can choose your own echo-chambers. With tiktok, the decision is made against your will and almost instantly.

My boring model of all this is just that there is such thing as "elites" and they have their own "elite culture". It sounds vague, but so are the effects we're trying to explain. There is no central authority at the top coordinating anything. The WEF is the non-profit think-tank version of any large progressive company. Internal signaling games are responsible for most of the sillier policy proposals (e.g. extreme covid measures, boycotting Dr. Seuss). The WEF may be more explicit in its intentions of changing policies, but it's not at all obvious that their influence is all that central in influencing elite culture. I'd be surprised if most elites had even heard of the WEF.

Can anyone steelman the case for any of the non-standard pronouns? Why hasn't the LGBT community settled on he/she/they, or even just exclusively using they?

Also curious what's the point of including the subject and object forms (e.g. he and him), seems redundant to me, unless someone is combining he/her or she/him? I've heard non-binary folks are doing similar that in languages where both the verbs and pronouns inflect based on gender and there isn't any neuter form (e.g. hebrew)

This isn't a novel claim in this space.

I'm not sure whether discrimination is the primary cause for differing outcomes between racial groups, but the question isn't as binary as it may seem. What if discrimination is responsible for the way that group's culture developed, or for selecting certain genes? It's been hypothesized that the type of discrimination jews faced for centuries placed a selective pressure for high intellectual achievement. If jews were pressured into professions requiring high IQ like doctors and banking (jews were the only ones allowed to charge interest in many places), then evolution would filter out the ones too dumb to practice medicine or banking. I'm inaccurately paraphrasing Scott Alexander's argument here but my point is that discrimination has downstream impacts that can last longer than the discrimination itself.

Interestingly: this argument actually makes me less anti-trans.

This is my view as well. I'm often accused of being conservative but there's something beautifully utopian about people just being who they want to be. It's a little messy today, but if technology were absolutely perfect and low-cost, who wouldn't try switching genders for a couple hours?

The concern isn't that tiktok is spreading pro-CCP or even anti-western propoganda. The concern is that it is addictive and stealing our youth's attention. Yes, other social media do this too, but tiktok is particularly good at stealing attention and time.

The chinese aspect comes into play when you realize that tiktok in china is a different app. If China thinks this attention-stealing is bad, they're going to fix their app rather than the international version. Also, tiktok allows them to China on international users (this isn't the article's point, but it is the main concern you hear about in mainstream media)

Sure, it's the same idea, but tiktok is much more powerful. You don't have to follow, like or even watch an entire video for the algorithm to respond. Being even slower to dismiss a video will boost similar videos. The youtube equivalent would be something like tracking your eye movement to see which thumbnails you're looking at. I know because I'm now getting porn-ish content after being slightly slower to dismiss videos with pretty girls.

Elite culture and universal culture have a lot of overlap, perhaps they're even the same thing, but it's certainly more concentrated and adopted within elite circles. In a typical company, employees express this culture proportionally to their rank. The elite culture gives you status, and you have to signal you're part of the in-group.

My model of Scott's universal culture is a natural common-denominator. Elite culture is more forced and over-the-top, due to the status it gives its members. Perhaps elite culture is downstream from universal culture.

Baristas and mic-girls might express the same attitudes on some social issues like gender and the environment, but different views on economic issues.

There are a lot of relationships in biology that generally hold across species but not within a species. For instance, mammal size is associated with longer lifespan across species, but intra-species it doesn't hold and can even be the opposite, e.g. in dogs.

Many comments are pointing out the variance between people, but considering frequent bathing is a pretty recent trend, I would think we'd get used to lower frequency bathing if we really tried (or were forced to).

Is anyone aware of evidence for some kind of feedback loop between bathing and oil/sweat/odor production? I know of many anecdotes of people who went from washing their hair everyday to much lower frequencies, they all say they got used to it and their hair doesn't get gross anymore (or as quickly). Personally I used to shower everyday and used to feel pretty disgusting starting around the 16th hour, but now I wash every 2 days and I feel fine until around the 40th hour. It could be psychological rather than physiological, but my hair really does look less oily for longer since I cut my bathing frequency. My guess is that it's a combination of psychology and physiological feedback loops (e.g. decreased sebum production).

It's also getting harder for women to find mentors because it's riskier for men to enter that kind of relationship. I don't have any ideas, I suppose mentors can do their best to vet their mentee candidates for the risky behaviour, and mentees can find ways to signal low-risk.