@MathWizard's banner p

MathWizard

Good things are good

0 followers   follows 0 users  
joined 2022 September 04 21:33:01 UTC

				

User ID: 164

MathWizard

Good things are good

0 followers   follows 0 users   joined 2022 September 04 21:33:01 UTC

					

No bio...


					

User ID: 164

1x. If it's a video I'm actually watching, I want to watch it as intended. If it's a "video" where the point is that people are talking, I have it on in the background while doing something else. With my attention split between the words and the something else, I'll lose track of what's going on if it goes too fast, and I won't get bored at 1x because my attention is already split.

If it's a video that needs to be actually watched and it's too boring at 1x, I'll just stop watching and find something better.

Right, but barring some sort of bungling by the defense, or dishonesty, or extreme bias, or something going horribly wrong, it should be impossible for the same collection of evidence to meet the burden of proof to actually convict both defendants "beyond a reasonable doubt". It seems to me like the same set of evidence that convicted the first one, simply knowing that a jury convicted the first one, is itself a reasonable doubt on the second one. It's possible that the first was incorrectly convicted of being the mastermind and the second is actually the one, but it's reasonable to doubt it.

"But battles are ugly when women fight."

What does he mean here? This could either mean that women fight dirty and thus make the battle ugly, or it could mean that women having to fight means women getting wounded and killed and being forced to wound and kill others which is itself ugly. Or it could mean that only in a most desperate and ugly battle for one's very survival do we forgo our principles and make women to fight because we need every last body at any cost.

Technically they could and should do it on their own, but clearly they don't, so some sort of handholding might help. Like, if the court could formally submit the law to the legislators with the specific part highlighted as being ambiguous or exploitable, and the case motivating this as an example, and then the legislature must handle it in a timely manner to clarify their intent. Given them an option to say "we endorse this exactly as written, even in light of the example you've submitted with it, the law needs no changes", but if they have to explicitly vote for that publicly then it removes a bunch of the status quo bias that lets these loopholes exist for so long. And also have a system where if the legislature ignores the bill entirely or can't come to a consensus on it then the court publicly admonishes them for being incompetent or something symbolic that doesn't give the courts any actual power beyond what they already have, but makes the legislators look bad publicly when they make stupid laws.

And limit the court's ability to do this to a certain number of times per year or per law or something to prevent activist judges from just harassing legislators with laws they disagree with. But I mostly just want a fastforward button on the existing "law runs into trouble in court -> legislators notice and care -> legislators fix law -> courts can do sane things without having to abuse language" process. And then maybe use this as an excuse to crack down on courts abusing language and bending laws towards what they think was intended instead of what it literally says. Just always do what it literally says and if you think the intent was different ask the legislature to make them match.

This is precisely why I think courts should have a "return to legislator" option, at least for future cases if not for the current one. Rather than have a war between the letter of the law and the spirit of the law and being forced to either ignore the spirit, or mangle the language of the letter until it fits what you think the law should be, this should be a case where the judge goes "Hey, lawmakers, we found a bug in your code, fix it!"

Now, it would be unfair for the people involved here for the law to suddenly change on them mid-court battle, and their own case should be decided based on the law as written, but the law could update for future cases.

In this particular case, the obvious solution is to adjust the "single-family" zoning law to prohibit renting to more than a single family at a time. You can build your 24 room "single family house", good luck finding single families that will pay you enough to recoup your costs when renting in series instead of in parallel.

You can do that, but it's rare. Therefore, as in Scott's Be nice at least until you can coordinate meanness (which I only partly agree with), it doesn't really do much unless it's a norm. The curve of behavior as a function of consequence is highly nonlinear, and a rule which is enforced in 0.1% of cases might as well not exist, since people will just ignore it. A world in which each instance of hooking up under false pretenses carries a 0.1% chance of getting beat up or having to go to court leads to basically no change in behavior and just increases violence to no benefit. It's only if men seeking to take advantage of women expect to actually face consequences, and have either had it happen to themselves in the past or to people they know, that they will factor those consequences and rethink their behavior.

Back in the day, fathers and brothers would take it upon themselves to defend the woman's honor. If a man slept with a woman under false pretenses of a long-term relationship and then just abandoned her, they would beat the crap out of him, ostracize him, and possibly even kill him.

We can't do that in modern society and, while the rule of law is useful and helps protect people from threats of violence for less significant offenses, I think something was lost here.

All of these are plausible depending on surrounding factors absent from the example. As written and inferring from base rates, and assuming some degree of mental illness on John's part, I'd think 1 is most likely, but 2 is fairly plausible, if we assume the mental illness is something along the lines of "pathological liar" that keeps him exaggerating even as it starts to fail to gain sympathy.

It could also be a more steelmanned version of 1, what if people do keep trying to kill him? Like, not these specific people at his work, he's still exaggerating about them, but what if it's other people? Maybe John lives in a really bad neighborhood and gets mugged once a week, barely escaping by throwing his wallet and running the other way. Maybe he has to fight tooth and nail and ends up in the hospital regularly badly wounded. Maybe it's not about John at all, it's just a really bad neighborhood and everyone who lives there gets mugged regularly, or maybe John looks like an easy target. But the repeated trauma makes him think it's a conspiracy and he's not smart enough to pick out the pattern: person in dark alley = mugger, person in office = friendly, and he just thinks all people have a 50-50 chance to attack him.

If we reduce his pattern-matching abilities even further, maybe he never actually gets mugged, but he keeps doing something stupid like climbing rusty fences and scratching himself, or drunk driving and getting in accidents that almost kill him, and generalizes that to people trying to kill him.

Or maybe he just watches too much TV and movies and people are trying to kill each other all the time (especially trying to kill the protagonist) and he thinks of himself as the protagonist, therefore people must be trying to kill him.


If instead, we increase his pattern-matching abilities, maybe he does regularly get mugged, or his friends and family members do, and he notices that most of the muggers in his bad neighborhood have a certain ethnicity, and so he becomes a racist. Or maybe he goes to the police to fix the issue but they don't take him seriously because he sounds like a paranoid nutjob (when he accuses the actual mugger and Alice from work in the middle of the same rant the police can't tell which one is real and which is exaggerated), then he becomes anti-cop, or anti-government, or anti-whoever is in charge of making the cops be so lax on crime and oh hey have you read this article about how such and such group is secretly controlling the government to be soft on crime or whatever?

Stepping out of the metaphor, which I think is somewhat of a weakman for this phenomenon, I think this simultaneously explains a large chunk of racists (in all directions), anti-religion, anti-capitalists, etc etc etc. Bad thing happens to person or to people that person knows, or hears from (sometimes signal boosted and exaggerated by the media, sometimes by word of mouth). There is a real pattern causing it to happen repeatedly, though sometimes it's a pattern as complicated as "The Entire Economy", it gets oversimplified, exaggerated, and then attributed to a particular group, and the people who believe this explanation become radical anti-that-group. It's a combination of paranoia and actual pattern recognition, because there usually is an actual legitimate instigating factor that is genuine Bayesian evidence against that group, it's just much weaker than would be needed to draw the exaggerated conclusions they come to. There ARE evil racist white men trying to keep minorities poor. There ARE worthless degenerate minorities who live on crime and welfare and contribute nothing to society. There ARE corrupt police officers abusing their authority. There ARE pedophiles in government jobs. There ARE Zionist supremacist Jews who want to control all of America and manipulate it into being pro-Israel. All you need is for someone to encounter some of these in real life, or evidence of them existing, and then the pattern matching can begin until it spirals out of control.

And some of these people will have genuinely convincing evidence on their side, by sheer random chance. 1% of people will be in the top 1% for people who have been mugged. 1% of people will be in the top 1% for people who have been unfairly harassed by police. 1% of people will be in the top 1% for people who have been stared at suspiciously by shopkeepers despite doing nothing wrong. 1% of people will be in the top 1% for people who have been laid off by a Jewish boss. They're going to look at the evidence they've seen with their own eyes and be unconvinced that it might be a coincidence. It doesn't seem like a coincidence, it seems super unlikely. If their life were admissible as a scientific paper it would reject the null hypothesis. p < 0.05. They're Jellybean people!

I think that's what a lot of this is. People who perceive patterns where there are none, people who pick out genuine patterns and misattribute them, and people who have coincidences happen to them that are rational evidence when viewed from their individual perspective but don't stand out when you adjust for multiple comparison tests.

It is material in-so-far as it modifies their branding/camouflage, tactics, targets, methodology.

It's not "just" corruption. A cop explicitly asking for bribes in America is going to get fired and prosecuted instantly. There are a lot of things cops can and do get away with, but explicit bribery is not usually on the list, with rare exceptions. But if they camouflage it, it's more possible and likely. Call it "asset forfeiture", make excuses for why it doesn't count as theft, and then target poor people with drugs that don't evoke sympathy from the public. It's still a form of corruption, but it's a very different form of corruption than the DMV official demanding an extra $100 to expedite your paperwork or it will get lost for 6 months, or a police officer pulling you over and threatening to write you a ticket for made-up charge unless you slip them $50. Just using the word "corruption" doesn't tell you the difference, or orders of magnitude, about what to expect to have to pay and how to avoid it.

Similarly, woke progressives are not going to pull over your car and extort $50 from you, nor will they arrest someone with marijuana and snatch all their money. They worm their way into industries and beurocracies and government agencies, then hire friends and family and like-minded people who look and act and speak like them, fire people who don't, and divert funds away from productive uses and towards spreading more woke-ism and getting cushy jobs for themselves and their friends. And they use nice-sounding ideals as a cudgel to choose who to attack and who to let go. I agree that this is a form of corruption, but it's a very different form of corruption than most other forms. And it's very very very good at masking the corrupt parts of itself and pretending to be good, and thus skates by unnoticed and in fact praised by so many people.

Calling it "corruption" is not wrong, but there's so much nuance it misses.

That's almost a prisoner's dilemma, but I suppose technically counts as Chicken. You're doing a (very slightly) negative sum interaction in order to siphon zero sum rewards away from other people. Nash equilibrium, everyone does this and ends up worse off than if they just respected the queue. (Although I suppose queues themselves are a bit of a prisoner's dilemma with respect to arrival time)

It reads like a joke that someone forgot the punchline to and butchered the delivery. Would have been much better if they had figured out how to get it in the correct order.

I don't think anyone has meaningfully decreased the power of the government in decades. Maybe a couple overreaching laws here or there got repealed, but plenty more came along, and the government just keeps doing whatever it wants with whatever justification they can make up ad hoc to justify the thing they already decided to do.

"Destroy your opponent before they can destroy you" does not at all sound like the "reasonable answer". Especially since this won't literally destroy them, they'll still exist and be even more ravenous to seize the reins of power. It seems like the actually reasonable answer is to de-escalate and decrease the power and influence of the government so people can make their own choices about their own personal lives.

I don't even get why there are "sides". I don't care whether the meat I eat comes from a "farm" or a "lab", I just care whether it's cheap, tasty, and nutritious. Let them both try their best and we can judge them and eat them according to our own preferences. I'm on team freedom, and that means nobody gets shut down pointlessly just to "own" the other side.

Precisely. The government's one and only legitimate role in this would be to mandate that it has to be labeled correctly and can't be falsely advertised as ordinary meat. Other than that, they need to stay out of it and let the people make up their own minds on what they choose to purchase.

It's unclear what he'll use as a replacement for it in the likely event that Trump sits there next year

Does he need a replacement? If Trump stays on the same side, at least with respect to the culture war, then there won't be anything he does that Texas would feel the need to nullify. Any overreaches of Presidential authority that Trump makes will probably be in favor of the right and against the left. At which point left-leaning states may try to pull the same stunt using these incidents as inspiration/justification. But as far as Texas is concerned they'll probably just cheer on whatever he does as something they were doing anyway or would like to do.

Seems like this depends entirely on what aesthetic the game is aiming for. Would The Matrix have been a better film if it were in the style of a Miyazaki film? Or as an animated version of a Leonardo da Vinci painting? From a certain perspective you might consider the Mona Lisa to be "better art" than The Matrix, but being real people with some minor CGI added worked for what it was going for and made a fantastic film.

We're not there yet, but if we can reach the point where hyper-realistic faces in games can actually be mistaken for real actors, there's a huge range of applications where it would actually be useful and better than a stylized game character because it fits the aesthetic the game intends, even if such situations are the minority.

Small-scale shower thought, since I don't want to wait until Sunday

You do realize that you're allowed to post in the existing thread, right? It still exists, it doesn't expire on Monday, it expires next Sunday when the next thread comes out.

I've been doing it like that, where they're all together and reference each other, it's just that then when Agent has 15 methods because some of them are experimental variations on each other or niche things I wanted to do to see what would happen, then I make another class for graphing scatter plots, and I've got a bunch of methods for (Make a world, then modifier the parameters according to X, then execute Y, then graph the results, then repeat that N times) that would be nice to stick in their own class somewhere, and then I've got a bunch of useful static methods that do stuff like load and save data to CSVs that would be nice to have in their own class for organization purposes. And if I just lay them out linearly (which I mostly have, with a few rare exceptions that definitely have 0 recursive dependencies and I actually have moved them to their own .py file) then I have literally 2000 lines of code I have to scroll up and down just to find the right class whenever I want to check to see what the name of the method I want to call is or something, and then scroll back down to find the spot I'm working on.

Do you know if there's a way to.... I'm not even sure what the right language is here.... put different classes in different .py files, or at least different tabs, without running into recursive dependency issues.

Like, in Java, I can make a World class that contains a population from the Agent class, and models an epidemic going through them, and the Agents have a bunch of methods internally regarding how they function as they get infected and recover and stuff. And if I pass a copy of the main World to each Agent when it's created, then when they do stuff in their methods they can call back up to the World, usually for counting purposes, they say "hey I got infected, increment the total infection counter" or "hey someone was going to infect me but I'm already infected, increment the redundant infection counter".

As far as I can tell, in Python I can't do that nicely. If the World class imports Agent, then the Agent class can't import World. I can resolve this by defining both classes in the same .py file, but then all my code is arranged 1-dimensionally and I have to scroll through tons of stuff to find what I'm looking for (or use ctlr F). Whereas in Java each class has its own tab, I can open or close or switch to, so well-behaved ones that I'm not working on don't take up space or get in my way. I'm not sure if this is a Python issue or just a Eclipse issue. Is there a way to split a .py file into multiple tabs so I can organize better?

Partly a response, partly hijacking this to ask a question of my own to everyone else: what are you using as a editor/compiler?

I programmed exclusively in Java for years, but my new boss wanted programs in Python so I've been doing that this past year. Using Eclipse, which is wonderful as an editor, since it lets me organize everything and highlights typos that I make and stuff.

Aside a whole lot of friction involving different conventions and abilities, I was annoyed that all of the Python editors people recommended seemed way less functional until I discovered that I can program Python in Eclipse if I do the right stuff. So I've been doing that.

I'm not sure what the general consensus is, because I'm mostly self-taught and program on my own, making mathematical models for research purposes that nobody else has to use or collaborate with, so I've probably got all sorts of weird habits that would make more sophisticated programmers cringe. So I can't tell how much of this is objective and how much is just me being used to Eclipse for so many years and having little experience with anything else. But I tentatively recommend looking into PyDev for Eclipse, because in my opinion it's nice.

In the original construction, you win if you choose the absolute best partner, you lose if you do not, so the 75th percentile is a guaranteed loss, no different from the bottom 1 percentile. You only want to maximize the probability of getting the actual best, so it has to be better than anything you've seen so far or there's no chance and no point settling.

However, you are right that in this modified version trying to maximize utility this no longer applies, and a proper optimal strategy should probably be a function f(n,d) describing what percentile you're willing to settle on as a function of what time step it is (n) and what your estimate of the distribution is (d), depending on what you've seen so far and your meta knowledge.

It seems to me like the best way to model this would be to have some multiplicative scaling factor on utility that diminishes over time, since 50 years of your life with the second best suitor is (probably) going to be better than spending 20 years of your life with the best suitor. Perhaps a linear decay to simulate amount of lifespan remaining, so the utility of choosing the nth suitor is their actual quality multiplied by (1-0.01n). Or maybe weight it more towards the beginning to account for youth and childbearing years, like (0.99)^n or something.

My guess, extrapolating from my own similar albeit weaker feelings, is that it's something along the lines of corruption of virtue. Taking what should be positive qualities and wasting them or subverting them. Similar to a beautiful delicious barbecued pig falling into the dirt and feeding the flies and germs instead of a person. Turning from something nourishing and healthy to people into something that only feeds rot and disease.

That is, kindness has the potential to do good things and have value and make the world a better place. But misplaced kindness that rewards bad behavior will incentivize them and make them more prevalent. The vast majority of toxic behaviors we see in society are there not because the people who do them are stupid people who can't figure out how to be kind, but because they have been systematically rewarded for those behaviors in some way or another. Being toxic works, at least on some level for some people. To the extent that toxic behaviors are analogous to disease, systematically feeding them with misplaced kindness is comparable to someone going out and feeding expensive high quality meat to a wild rat colony, or a beautiful woman allowing mosquitoes and hookworms to feast on her blood. Not only are they wasting something good, but they are actively allowing badness to reproduce for everyone else to have to deal with. It's disgusting.

I feel less scorn and revulsion for people who do this accidentally than I would for someone doing this on purpose. And less for people doing this on a social level than with literal diseases. But it's still gross in a similar way.

1: Outlaw abortion, provide free non-abortive birth control for everyone, which is mandatory for heterosexuals until they get married, or at least sign some sort of contract with their long-term partner about how potential children will be raised.

2: School vouchers, which includes subsidies for homeschooling, provided said schools and homeschools pass certain accreditation procedures so they are actually providing a legitimate education and aren't just scamming for money.

Actually.... you asked for things out of left field, and 2 seems too normal. So to play more into the thought experiment here, let's ramp it up beyond what I'd normally advocate. Massively subsidize homeschooling. Provide free teacher training for parents, with a specialization towards teaching in small classroom sizes of children who you get to see repeatedly year after year, in every subject, and teaching kids to teach themselves. Pay enough that families can go from living on a two-working-parents income to a two-working-parents-but-one's-job-is-homeschooling-their-own-kids income. It won't even cost money, just divert it from the public schools who don't need as much because many kids have moved to homeschools. Have economists figure out every possible way that bad-faith-actors could exploit the system and figure out how to prevent it.

In terms of scale it involves more people, but in terms of perceived threat and actionable measures it seems less threatening.

Like, JFK was assassinated. This is immediately violent. Believing that the government/CIA assassinated the president makes them dangerous bad guys who are willing to assassinate people they don't like, and potentially justifies violence against them in retaliation and/or self defense. 9/11 likewise killed lots of people, making the perpetrators dangerous and worth retaliating against (even ordinary non-conspiracists can get behind this, which is why there was so much support for military intervention in the middle east after 9/11).

The most likely response to threats of violence are accumulating weapons to defend oneself and possibly pre-emptively strike using violence. If someone points a gun at you, you point one back.

Vaccines and Flat Earth are about scientific lies. They say that the leading scientists and media are corrupt and in the pocket of the government or whoever is leading the conspiracy, and the things they say cannot be trusted. Nobody needs to die to cover up the truth, because they can be paid off instead. Now, maybe some of the variants of vaccine and Flat Earth conspiracy theories do involve the government murdering people to cover up, and those ones are potentially dangerous, but I have never heard a Flat Earther talking about assassinations, so I think it's uncommon.

The most likely response to media and scientists lying is to not trust them, and possibly have this mistrust bleed into other domains. If they're lying to you about X, why should you trust them about Y? Now this can lead to some harms such as people refusing to vaccinate themselves or their children, but this is significantly less dangerous than actual violence. If someone lies to your face, you lose respect for them and possibly try to avoid them, but very few people would respond with violence (except in weird edge cases, where it's probably not about the lie itself but about the underlying thing they were lying about).