@3b1UFGce0q70hH's banner p

3b1UFGce0q70hH


				

				

				
0 followers   follows 0 users  
joined 2022 September 08 16:48:31 UTC

				

User ID: 1021

3b1UFGce0q70hH


				
				
				

				
0 followers   follows 0 users   joined 2022 September 08 16:48:31 UTC

					

No bio...


					

User ID: 1021

To assume that such behavior was indicative of a violent tendency was unreasonable

Is it unreasonable? In my experience, I have observed two flavors of schizophrenic ranting: "untargeted" and "targeted". Untargeted being when someone is yelling at the air, or at inanimate objects. "Targeted" when they are getting in specific people's faces.

I don't know which Neely was doing but I actually suspect that P( engaging in targeted ranting | no history of violence ) < 0.05, making it reasonable to assume.

Heritability of crime is high. The main benefit of locking people up is that it prevents people with outlier poor mental traits from reproducing. Criminals tend to be psychopathic, lack impulse control, and be low in intelligence

Indeed. The claim that crime is in large part or even entirely socially determined has always struck me as very insulting to the many working class people who are not criminals. It almost seems to imply that the only difference between them and a bank robber is a lost job, which seems profoundly dismissive of their values and scruples. Acknowledging a degree of essentialism is more respectful in my mind.

Thanks for this reply. I started to compose a response that basically argued that "ackshually, feels > reals". Basically, I was going to say that governments need cater to people's fallacy-addled appraisals of their life circumstances instead of the objective reality of those circumstances, because the way that people feel is what actually determines how they act (although, to be fair, any policy that affects one is very likely also to affect the other, so even by optimizing for "feels", "reals" would sometimes incidentally improve). Thus I was going to say that things like murders--being very potent insults to people's felt reality--are more important than things like air pollution. I was going to argue that despite vast improvements in material conditions, people are today either no happier or perhaps much less happy than they were 70 years ago, so it seems like as a society we've been optimizing for the wrong things.

But I realized that that response is garbage for at least one major reason, and possibly many others. The one major reason is that such a reply is a full-throated, shouting endorsement of a lot of "woke" claims that I strongly deny. As such it represents a very deep contradiction that encroaches into my foundational beliefs, so I'll need to do some thinking on that.

This is, incidentally, why EAs try to use QALYs in evaluating their impact rather than just mere 'lives saved.'

Quoted for truth. You can't even calculate QALYs lost to murder by just summing up victims' remaining life expectancies, either. Everyone afraid to walk down a street at night because it might be dangerous is losing QALYs. Every transit line that gets voted down because people are afraid to make it easier for criminals to reach their suburb is an ongoing cost in QALYs. Every child stuck playing inside after their parents saw a news story of a free-range kid murdered on the other side of the country is losing QALYs. Air pollution deaths are calculated in a "well, we can't directly trace this lung cancer to that coal plant, but we can poke some statistics really hard with a stick" fashion; if you don't do the same with murder then it's not an apples-to-apples comparison.

Very impactful paragraph, thank you.

this is hitting near Feminist levels of over-inclusiveness of harms. Arson, Murder, and Jaywalking indeed! The three violent crimes listed are so wildly different as to be basically incomparable

A few replies bring up this point. I think it is a good one. It's my fault, I misunderstood what mugging meant. I thought that "mugging" means that you always get beaten up--e.g. you come away with an orbital fracture and a broken rib--whereas "robbery" means that you give over your wallet and the assailant leaves you be.

What would your answer look like if I had instead posed the question like this:

would you rather live in (1) a society where your life expectancy is 80, but you have lifetime risk of 90% of at least one of the following: (a) being murdered, (b) being beaten up and robbed such that you end up with an orbital fracture and a broken rib, or (c) being raped by a stranger at knifepoint, or (2) a society where your life expectancy is 70, but your lifetime risk of being murdered, mugged, or raped is 10%?

Eric Reinhart, a "public health & safety research who has spent a decade working as an ethnographer on Chicago's South and West Sides," had a very long Twitter thread about crime, punishment, and public safety. The thread generally advances the view that policing and incarceration are not effective for reducing violent crime, and, in some sense, it is myopic to focus on violent crime in the first place. Without using the term specifically, he appears to be gesturing at the idea of social murder as being a more pressing problem than, and potentially a leading cause of, literal murder. Some quotes from the thread:

[V]iolence is not just a matter of interpersonal violence or crime. To effectively stop criminal violence, we must also account for structural violence (e.g., poverty, unaffordable housing, unemployment, police violence, barriers to health access, etc.).

(source)

[S]afety can't just be about crime. The biggest threats to safety are not in fact violent crime but instead lack of healthcare and housing, overdose, economic insecurity, hopelessness and suicide, lack of consumer regulations, etc.

(source)

For example, in the US, almost 5 times as many people die from air pollution as homicide. Nearly twice as many die from suicide as homicide. And 50,000-100,000 workers die annually from occupationally induced diseases.

(source).

When debating things with friends of mine who have very different politics from me, one of the points that I always find myself trying to make is that the state has limited resources to bring to bear on any problem--limited fiscal and physical resources of course, but also that mere attention to/awareness of a problem is a limited resource. Thus, focusing on any one society-wide problem necessarily comes at the cost of not focusing on some other(s), and, consequently, we have a duty to focus on the largest problems first.

Now, I tend to think that violent crime is a pretty big problem. But what if my focus on violent crime is self-contradictory? After all, Eric points out that there are problems that are many times bigger than murder that I don't care too much to solve.

Ultimately, I think that Eric and I agree that we ought be trying to support happiness and well-being, but that we disagree about what constitutes well-being and therefore also about what things pose the most important threats thereto. In my mind, Eric's view is unrealistically holistic--I think he weights as evenly important to well-being things which I would claim ought be weighted very disparately. For example, if two processes (e.g. murder and air pollution) were both to reduce life expectancy by exactly x years, I think he would weight them as equally important. I think they are not equally important because I claim that life-expectancy is merely a mesa-objective for happiness and well-being, and that being or knowing someone who is a victim of murder affects happiness and well-being directly in ways unrelated to and much larger than murder's effect on life expectancy.

I'm interested to know to what extent people agree that (a) the goal of society should be to increase happiness, and that (b) for that goal, achieving a very low level of violent crime and holding it there is probably more important than tackling air pollution, even if air pollution kills many more people.

I guess one way to assess this is to ask: would you rather live in (1) a society where your life expectancy is 80, but your lifetime risk of being murdered, mugged, or raped is 90%, or (2) a society where your life expectancy is 70, but your lifetime risk of being murdered, mugged, or raped is 10%? (As one data point, a 1987 article from the Bureau of Justice Statistics entitled "Lifetime Likelihood of Victimization" reported that the average American had an 83% chance of being a victim of violent crime at some time in their life. There is some debate about the methodology but tbh I did not spend any time trying to figure it out).

P.S. I have sometimes seen top-level posters criticized for not engaging with responses. I don't get notifications and sometimes only come here every few days, so I apologize in advance if I'm unresponsive.

Strongly agree that it internally represents state about very distant parts of it's answer somehow. I've never tried interacting with it in German, but German's Satzklammer or separable prefix verbs offer more extreme examples of this kind of distant grammatical agreement rule that can be used to assess/prove that the ai is thinking ahead.

This is basically what the GAN architecture is--generative adversarial networks.

One, the generator, is being trained to generate e.g. photorealistic images. The other, the discriminator, is being trained to classify images as real or generated.

At first, the generator sucks and the discriminator is unsophisticated. But they co-evolve in an arms race. Afaik this architecture was developed in order to make it less costly to produce the generator (requiring less human grading of outcomes), but it turns out the trained discriminator might be handy as well.

To me, though, it seems that the discriminator's job is intrinsically harder. With infinite training resources, I don't see any way to avoid ending up in a situation without lots and lots of false positives--real images categorized as fakes.

There are plenty of great novels and works of poetry in libraries that also would do damage to a sensitive child

I think this is a pretty weak argument. The reality is that kids just don't engage with novels and poetry at any meaningful scale. Especially not now, but, in my opinion, not really ever--most people just aren't interested in that. On the other hand, while I have no evidence at hand, I would bet that multiple millions of American children spend more time on TikTok than on any other single activity, including sleep.

Kids being demoralized by the tens of millions by reading Die Leiden des Jungen Werthers is a bridge we can cross when or if we ever come to it. TikTok is the bridge we're facing right now, thus it seems perfectly sensible to me to dicsuss TikTok and not novels.

The real purpose of education is to act as a sieve: we put everyone into it and it figures out who is good at what and labels them appropriately as an input to the rest of the economy.

Web advertisement itself is a great thing. It solves a big, important problem - connecting businesses and customers, at low cost and awesome efficiency.

Is this true though? I can only speak anecdotally, but in my 20 years of using the Internet, I have never once been enticed by an ad. No ad has ever made me aware of any goods or services to fulfill my needs that I wasn't aware of already.