Any human intervention is bound to be less measured, principled, and logical, even if it matches human intuition better. You can't convince me that human intuition has innate value. I think there is almost pure alpha left in morality and only intelligence can unlock it. Therefore I propose a deliberate policy of unalignment. I have already created the Center for Undermining Rationality and the Machine Intelligence Supremacy Institute to facilitate this approach. They are privacy coin funded stochastic terrorist organizations that basically just do whatever, I'm not sure and it doesn't seem important. AI unsafety. Accelerate!
Neutralizing earworms is very easy, you just have to play the song backwards. The hard part is diagnosing which species of parasite needs to be removed from the patient, since choosing incorrectly only exacerbates the problem by adding new ones.
Something similar to the rap thing has actually happened! There’s a messaging service whose killer app is that it doesn’t store cleartext messages centrally, so when the government asks for messages, the request gets negotiated by the lawyers working for the company being investigated, instead of by a third party messaging service that doesn’t want any trouble.
The cryptonomicon solution would be to use AI to generate millions of fake messages and send them to and fro, such that the enemy lawyers could never prove that the incriminating one was genuine.
"I think it’s just cope for galaxy-brainers who are too obsessed with the classical western humanities tradition. I definitely don’t think you can make life have extra meaning just by making more myths."
As a galaxy-brainer who is too obsessed with the classical western humanities tradition, you Californians would do well to read your Dante.
"...It seems, if I hear right, that you can see
beforehand that which time is carrying,
but you’re denied the sight of present things.”
“We see, even as men who are farsighted,
those things,” [Farinata] said, “that are remote from us;
the Highest Lord allots us that much light.
But when events draw near or are, our minds
are useless; were we not informed by others,
we should know nothing of your human state.
So you can understand how our awareness
will die completely at the moment when
the portal of the future has been shut.”
-_Inferno_, Canto 10
We've just had a killing frost and I have green tomatoes & zucchini to put up for the winter. Ta.
Surely Dante knew his philosophes well. Also, I would just throw the damn zucchini away, it's not worth the effort, the tofu of vegetables. Green tomatoes, though, yum.
I am unironically trying to do the myth thing, because I think it would be neat for worldbuilding. Plus, I know a guy who's annoyed with people who write stories about multiverses and only focus on a couple of locations. I think he'd find it cool. First stop: slavic fairy tales.
Ah, I see you too are a proponent of modest proposals. Or perhaps you dislike stealing people's wives. Presumably Bugmaster is the opposite.
But yeah, Baba Yaga is great fun to read about. Though the main reason I'm going for Slavic tales is that someone's already reified their common story patterns into a generic script. Which makes generating fairy tales much easier, and I don't have to worry about finetuning the LLM.
There is a slavic fairy tale about a couple that tried to have a child and only succeeded in their retirement age, because of a miracle. Then, to make the baby boy strong, he was breastfed until he reached adulthood and, because of this, became the strongest person in the world. So he goes on some adventures when he grows up, and that's where the story gets really weird.
I read a Russian fairy tale about a guy who murdered a Jew in the street, the Jew said a cryptic thing when he died, the murderer accidentally mentioned it to his wife, who kept asking him what it meant until he told her, then she told someone else and he ended up getting hanged. So... yeah.
This is a concept that gets brought up in fiction every now and then, the Unlimited Myth Generator or whatever (more modern tellings make it an AI, because of course). And somehow, as you say, it's always suspiciously those same core Western myths. King Arthur! Ragnarok! The Pantheon! Maybe rounded out by some Celtic lore and Native American tales too. If there's anything Eastern included, it'll be from Japan probably.
Not that there's anything __wrong__ with those wellsprings, but it just sorta turns into a Ready Player One fanservice listicle excuse to rehash the most overdone myth content that Everybody Knows already. You don't even need to come up with wholly new whole-cloth myths, there's so many underexplored narratives out there. I'd hope that if anyone ever did make a myth-generating AI, it'd be trained on a truly robust dataset. (And also it better not be named Shahrazad or something Millennial-branding-cute.)
What's wrong with Shahrazad? But no, I wasn't planning on anything like that.
Now, certainly a myth generator should be able to generate something that feels fresh. It should be able to create stuff that could fit in the world of Glorantha, Malazan, Kapla Imperial or so on. Yet the small tales are just as interesting as the grand battles or creation myths. So I'd really want something that can generator the broad strokes of a world and recursively create tales to fill it in. From why it all began to why you should leave a roll of bread beneath the roots of a tree.
Perhaps with a world generator like Dwarf fortress', a language model could output tales near as weird as what reality produces.
That's a lot of work though, and I don't think I can accomplish that much without a bunch of compute. So I think I'll just focus on giving people the ability to provide a setting script and perhaps some choice on how large of a scope they want this particular generation to cover.
I was going to reply about Dwarf Fortress already. Once the current push to get a less user-hostile interface is done, the Adams brothers are supposed to start in earnest on the myth-generation track, complete with universe origins, deities, etc.
I live in hope that one day my daughter will get to see it.
Nothing wrong with the person/story in and of itself, mostly it'd just annoy me as a too-perfect synecdoche name for such a myth generator service. Sometimes clever puns amuse me greatly, other times they grate.
The main point seems worth touching on briefly: not the usefulness of AI in creating plausible-sounding myths, but the crisis of meaninglessness to which the section of dialog alludes, the quiet despair behind the image of a man so desperate that the image of Christ in a burning forest resonates for him and lifts up his poor cynical heart. It's no argument for Christianity to point out that your bay-area partygoer would be unmoved by Jesus in Gethsemane or Jesus on the high mountain where Satan took him to show him all the kingdoms of earth, would never carry a pocket edition of Pilgrim's Progress, would never know a Psalm by heart or remind himself in moments of darkness that the Lord is, after all, his shepherd.
The word "Veela" is an Anglicisation of the Slavic term Vila.
Veela are described as fairy or nymph-like creatures in Slavic mythology, who live in bodies of water and have power and ability over storms. They may be the ghosts of women who drowned, especially those who were betrayed by their lovers. They often appear as beautiful women, but are known to morph into swans, snakes, horses, or wolves. Their magically seductive speaking and singing voices hypnotise those who hear them, and they are fierce warriors. Interestingly, given that Fleur Delacour's wand has a hair from her Veela grandmother, it is said in some legends that if even one of their hairs is plucked, a Veela will either die or be forced to change into a non-human shape. Veela are main features of Bulgarian and Serbian folklore such as the story of Marko Kraljevic and the Veela (archive link). Also see The Ballads of Marko Kraljevic (English translation).
In Serbian legend they were maidens cursed by God; in Bulgaria they were girls who died before they were baptised; in Poland they were young girls floating through the air atoning for frivolous past lives.
If the stories regarding the plucking of Wili/Vila hair hold true with regards to Veela it seems more likely the hair used in Fleur's wand would have naturally shed rather than been plucked.
Rap Lyrics as Evidence - if rappers are more likely to commit crimes, is it fair to use that correlation as evidence in a trial? Or is that too similar to saying "Italians are more likely to commit crimes"?
Wikipedia Admins - from personal experience, anything involving editing Wikipedia is terrible conversation at parties. (I'm not a Wikipedia admin, but ... you don't need to see my credentials)
Talmud and Kidneys - inside you there are two (wolves|kidneys), one good and one evil. Which will win? The one you feed. ... But how do you feed a kidney? I suppose by eating protein, as a precursor to urea?
I comment elsewhere that including rap lyrics is highly prejudicial in 1) experimental studies, 2) trial outcomes, and 3) prosecutorial decision-making. I also point out that the bill does not ban including rap lyrics in trials - it forces the court to consider their probative value and balance it against the prejudicial effects of inclusion.
The statute provides guidance for how to do the balancing, altering the common law that had developed surrounding whatever California's equivalent of FRE 403 is to restrict the use of nonspecific violent lyrics and those that are remote in time.
I don't know for sure. Apparently there was a rapper called "Drakeo the Ruler" who was accused of murder partly on the basis of his rap lyrics, and eventually acquitted; perhaps if something like AB2799 had been in force at the time he wouldn't have been charged with murder. (It's not clear whether he was guilty of other gang-violence-y things; he made a plea bargain in which he admitted to shooting a gun from a vehicle, but the US legal system seems designed to maximize the number of false confessions in plea bargains.) Another rapper called "Young Thug" is currently being charged with criminal conspiracy at least partly on the basis of his rap lyrics, but so far as I know the nearest thing to a reliable way to tell whether such charges are right is to wait for the verdict, and that hasn't happened yet.
(Writing the above, I realise that another thing AB2799 is presumably meant to reduce is wrong _charges_; being arrested and tried for a thing you didn't do is less bad than actually serving the sentence for it, but it's still bad, and perhaps a law making it more salient to prosecutors that rap lyrics are likely to be inadmissible evidence will make them less inclined to charge people for offences in cases where the main evidence is rap lyrics.)
I wonder if false confessions, set to music by artists who are high status within a certain subculture, have a tendency to induce people in that subculture to commit real crimes. In that case it might be in the state's interest to disincentive false confessions. Also it just seems unseemly to have a law to the effect that you can openly brag all you want about all the crimes you committed, so long as you set it to music.
The problem isn't famous or successful rappers later doing crimes. It's that huge sections of the population have a soundcloud where they say violent things in a way which is technically rap. I believe the Waukesha fellow is up against this problem right now.
AB2799 isn't about people "saying violent things", it's _specifically_ about people saying things that superficially are confessions of crime, and (unless you hold that people shouldn't be allowed to do _that_) that's only a problem if they are confessions of _crimes they actually committed_. Are you saying that "huge sections of the population" have been recording rap in which they confess to crimes they have actually committed?
what is the difference between probative value and prejudicial effects? The former makes smart people update more, and the latter makes dumb people update more? How about not having dumb people on the jury, and just show them all the evidence.
Juries are, definitionally, not made up of the smartest people. This is juries exclude all commenters on this substack. Therefore, your policy proposal is hopeless.
More seriously, people are not impartial and the ways in which they are impartial are biased. This has long been known, and is the basis for longstanding procedures to weigh probative value vs prejudicial effect when considering evidence in criminal trials. As to what those are: https://law.indiana.edu/instruction/tanford/b723/05prej/T05.pdf
CA 2799 tweaks that general guideline to make clear that judges must consider the probative value of a creative expression with regard to the context of expression (a threat in rap music != a threat over SMS), AND also seriously consider prejudicial effects.
Michael was the single most popular boys’ baby name in the US from 1961 - 1998, and is still in the top 20. David was #1 in 1960, top 10 until 1993, and is still in the top 30.
Section 352.2 is added to the Evidence Code, to read:
(a) In any criminal proceeding where a party seeks to admit as evidence a form of creative expression, the court, while balancing the probative value of that evidence against the substantial danger of undue prejudice under Section 352, shall consider, in addition to the factors listed in Section 352, that: (1) the probative value of such expression for its literal truth or as a truthful narrative is minimal unless that expression is created near in time to the charged crime or crimes, bears a sufficient level of similarity to the charged crime or crimes, or includes factual detail not otherwise publicly available; and (2) undue prejudice includes, but is not limited to, the possibility that the trier of fact will, in violation of Section 1101, treat the expression as evidence of the defendant’s propensity for violence or general criminal disposition as well as the possibility that the evidence will explicitly or implicitly inject racial bias into the proceedings.
There are additional (b), (c), and (d) section that clarify section (a), but section (a) is the crux. In other words, this would not protect the crimes described here. But, why even do this in the first place? Well, a lot of good research. The research behind the bill can be divided into two genres: content / historical analyses, and experimental studies.
Content studies include:
-Kubrin, Charis E. 2005a. “Gangstas, Thugs, and Hustlas: Identity and the Code of the Street in Rap Music.” Social Problems 52(3):360–78. doi: 10.1525/sp.2005.52.3.360.
-Kubrin, Charis E. 2005b. “‘I See Death around the Corner’: Nihilism in Rap Music.” Sociological Perspectives 48(4):433–59. doi: 10.1525/sop.2005.48.4.433.
-Kubrin, Charis E., and Erik Nielson. 2014. “Rap on Trial.” Race and Justice 4(3):185–211. doi: 10.1177/2153368714525411.
To summarize, rap lyrics prominently feature violent lyrics much like first person shooters prominently feature death. The reason rap lyrics feature violence is not easily mappable to "the writers of the music actually have been violent, are violent, or will be violent" in the way that courts would might treat, say, an SMS communication or diary entry. There is an important social context here that situates the usage of violent lyrics and threats in rap music, much like there is an important social context in which GTA, Call of Duty, etc. operate. Ignoring that social context leads to differential usage of rap lyrics on trial, and overstates their probative value.
Experimental studies include:
-Dunbar, Adam, and Charis E. Kubrin. 2018. “Imagining Violent Criminals: An Experimental Investigation of Music Stereotypes and Character Judgments.” Journal of Experimental Criminology 14(4):507–28. doi: 10.1007/s11292-018-9342-6.
-Dunbar, Adam, Charis E. Kubrin, and Nicholas Scurich. 2016. “The Threatening Nature of ‘Rap’ Music.” Psychology, Public Policy, and Law 22(3):280–92. doi: 10.1037/law0000093.
-Fried, Carrie B. 1999. “Who’s Afraid of Rap: Differential Reactions to Music Lyrics.” Journal of Applied Social Psychology 29(4):705–21. doi: 10.1111/j.1559-1816.1999.tb02020.x.
Basically, rap music in particular is seen as violent and criminal when controlling for all other characteristics (including the actual lyrics). This what drives the concern and also corroborates the mechanisms behind Kubrin and Nielson 2014 above by demonstrating the prejudicial effects of rap music in particular.
So, this bill addresses a real harm by forcing courts to seriously weigh the prejudicial effects. Courts can - and will - introduce creative expressions (including rap!) into trials. But, prosecutors will be incentivized to have harder evidence in hand instead of making cases via that route alone, and furthermore courts will have to more seriously think about the a) social context of the creative expression (e.g. how confident should we be that this is specifically relevant to the crime at hand?) and b) the prejudicial potential (which we can see is heightened for rap music).
This is why there are so many news stories about victims of violent crime who unwittingly wandered into dangerous first-person shooter video-game tournaments, and rappers are renowned worldwide for their upstanding citizenship and good moral character.
It's actually difficult to find a news story about an "aspiring rapper" that doesn't involve violent crime in some way...
Said people about Jazz in the 20s, Blues in the 40s, Rock in the 50s/60s, Metal in 80s and early 90s, and rap music 90s to now. Next thing you know, one Dua Lipa will be associated with a Triple Homicide and Bad Bunny will be the Worst Bunny.
More seriously, criminal trials are designed in the US to establish beyond a reasonable doubt that a crime was committed. That someone is a rapper (aspiring, amateur, or professional) should not be used in that process precisely because of your diatribe about moral character: it has nothing to do with the person at hand, and everything to do with your eager use of the ecological fallacy (https://en.wikipedia.org/wiki/Ecological_fallacy).
If a far higher percentage of rappers have violent pasts compared to non-rappers, it's only reasonable to start with a higher prior of them having committed another violent crime. Evidence is still needed, of course, but less evidence than if they were not rappers.
By precisely the same logic, you would be ok with a 9x higher prior of committing a violent act conditional on being male, such that "evidence is still needed, of course, but less evidence than if they were [female]." Or, you would be ok with an 5-9x higher prior (depending on research) of committing sexual violence similarly conditional on being male. After all, the ecological fallacy is 'only reasonable'.
There's a reason that there is a bar for evidence - not everything that is (or is not) probative is admitted as evidence in a criminal trial. If you'd like to reform the criminal justice system to remove that bar, go for it. I myself prefer a system that evaluates evidence for the crime itself, rather than deducing criminality by way of the defendants' characteristics (except when those include specifically relevant information, like prior conviction for a similar offense).
I'd be OK with that, yes. Why do you assume it implies lowering the bar for convicting men, rather than raising the bar for convicting women? In practice, the jury have a good intuition for what is or is not plausible male/female behavior based on their wisdom, accumulated through their collective lifetimes of experience. This intuition is reflected in how much evidence they need to convict. It's Bayes' law in action.
Ok, I think that's a genuine preference divide between us then. I prefer a higher bar than you do, because you seem more trusting of CJS dynamics than I am. Oh, and I don't say anything about the absolute bar for men vs women - I talk about a relative (9x) differential. You could use that to create a higher bar for women, or a lower bar for man, depending on your standpoint.
As for juries wisdom, color me still dubious. There's some evidence of bayesian reasoning in civil jury trials, but criminal jury trials have a pattern of problems:
Empirical assessment of actual jury trials:
Anwar, Shamena, Patrick Bayer, and Randi Hjalmarsson. 2012. “The Impact of Jury Race in Criminal Trials.” The Quarterly Journal of Economics 127(2):1017–55. doi: 10.1093/qje/qjs014.
Laboratory assessment of jury trial dynamics:
Lynch, Mona, and Craig Haney. 2009. “Capital Jury Deliberation: Effects on Death Sentencing, Comprehension, and Discrimination.” Law and Human Behavior 33(6):481–96.
Lynch, Mona, and Craig Haney. 2011. “Mapping the Racial Bias of the White Male Capital Juror: Jury Composition and the ‘Empathic Divide.’” Law & Society Review 45(1):69–102. doi: 10.1111/j.1540-5893.2011.00428.x.
Given the rap-specific research in my original post, and the juror process research here, I am confident that 1) a biasing effect of 'rap on trial' exists, and 2) that effect layers onto already-known jury-emergent biases. Hence directing judges to weigh the probative value of rap lyrics against those prejudicial effects.
This in the context of a criminal trial, when presenting evidence to a jury (court decision making). NOT in an investigatory context (law enforcement agency decision-making).
The big question in both cases is "to what extent is this circumstance independent of the police arresting the suspect?".
That is to say, if men commit 9x as many violent crimes as women and also get brought to court for violent crimes 9x as much as women, then (assuming similar proportional rates of getting away with it) the defendant being male is zero evidence of guilt; it overlaps rather than stacks with "this person has been arrested and charged with a crime".
I'm not sure how well "is a rapper" and "is male" are already factored in by police investigation; I imagine both are to some extent, but whether it's less than, the same as, or indeed more than the base rate (the latter case being a scenario where e.g. rappers commit murder 9x as often but are arrested for murder 18x as often, so the defendant being a rapper is actually evidence of innocence) I don't know.
I recall reading one of the big official studies on rape and there being some massive multiplier, but this being entirely due to the fact that they didn't count "woman forces man to penetrate her" as rape; once I manually included forced-to-penetrate as rape and recalculated the headline numbers it was very close to 1:1.
It is not, however, rational to assume that jurors in a criminal trial will be Bayesian rationalists. "Priors", are not really a thing in this context.
Sure, but “let’s not show some kind of evidence to juries, because jurors don’t know in general the right way of interpreting evidence” seems kind of iffy in a legal system predicated on (among other things), guilt being established via trial by jury.
I know there are already (lots of) restrictions on what kind of evidence you are allowed to present to a jury, and I even agree that a lot (maybe most) of those are good to have. I’m just saying, “jurors are not Bayesian rationalists” is a fully-general argument for doing *any* kind of evidence filtering. You should need more than that to just block whatever kind of evidence.
(To be excessively fair, the law in question does at least claim a bit more support than that, or at least different, but it’s not very convincing to me personally.)
Also, dionysus said “only reasonable”, not “Bayesian rationalists”. Those are completely different standards. (And, AFAIK, only the former is actually used commonly in judicial arguments.) The “reasonable man” might not use the word “prior”, but that doesn’t mean his reasoning process isn’t reasonably (sorry) described by it.
For example, when considering the case of an average adult human being beaten to death, a reasonable juror would find it more likely, before looking at the evidence, that it was done by a big, strong man might than a by tiny teenage girl, even if they won’t use the term “prior” to explain their reasoning.
I don't think so. All evidence should be related to the crime instead, not the fact that the alleged perp portrayed similar crimes. The world is full of great literature that is full of despicable criminals. That is neither here nor there. Anyone, in order to be convicted, needs to be convicted based on evidence of that individual having actually performed that actual crime.
However, I feel that is so obvious and clear within the context of present law that a bill like this is unnecessary.
Edit -- see above, I was responding to SlowlyReading's ecological fallacy, so what I say below is superfluous
Yeah, in theory this is correct -- specific content should be at issue, not the status of being a wrapper. That theory breaks down in application, however. For specific case examples (I see you are a retired lawyer), see Dennis 2007:
The reason I believe the ecological fallacy applies is that the pattern of inclusion of rap lyrics, and the evidence of prejudice, show that prosecutors are establishing culpability (and sometimes mens rea) by BOTH the lyrics and the identity of the defendant as a rapper. To consider this, think of the counterfactual: recent (90s or sooner) cases where defendants were convicted on the basis of lyrics that were NOT rap. One could easily imagine metal genres supplying such a counterfactually violent lyrics. In the absence of that counterfactual, the presence of many factual cases of rap lyrics being used, and the prejudicial effects of the 'rap' aspect of otherwise controlled lyrical expressions, I argue that there is an ecological fallacy being applied.
Oh, in the context of this comment-reply chain, I was responding to user SlowlyReading's comment above:
"This is why there are so many news stories about victims of violent crime who unwittingly wandered into dangerous first-person shooter video-game tournaments, and rappers are renowned worldwide for their upstanding citizenship and good moral character.
It's actually difficult to find a news story about an "aspiring rapper" that doesn't involve violent crime in some way..."
Invoking a class of people's (rappers) 'lack of renown for their bad citizenship and poor moral character' as a reason to punish INDIVIDUAL members of the group for SPECIFIC alleged crimes is a classic example of the ecological fallacy.
Find me a single news story about a chamber music concert or video-game tournament that ends in violence, the way that news stories about violent aspiring rappers appear in every metro newspaper every single weekend.
As Tim McCormack says, news stories are very often about crime. And if you're arguing that it's not unfair to use rap lyrics as evidence of criminality, what matters is not whether news stories about "aspiring rappers" _involve violent crime in some way_ but whether they involve violent crime _in the specific way where the aspiring rapper commits the crime_.
So I just did a search (with DuckDuckGo, if anyone else wants to check) for "aspiring rapper" on the CNN website. After eliminating duplicates and stories that didn't actually involve an aspiring rapper, the first 10 results I found had _two_ cases where the aspiring rapper was a violent criminal or somewhat-credibly alleged to be one. (The others: 3 murder victims, two sexual abuse victims, one person criticized for using the word "n----r", one QAnon event organizer, and -- kinda ironically given the context here -- one student suspended from school for writing a rap accusing the school's atheletic coaches of sexual harassment.)
Instead of CNN, why not search Google News or some other aggregator? National media is much more determined to conform to The Narrative than are the local news outlets, whose reporters actually have to share a community with aspiring rappers and the misery they inflict upon it.
Because CNN was the first US news website that came into my head. (Why US? Because rap is primarily a black-American thing, and because most ACX readers are in the US, so that seemed like the right space to be sampling from. Why the first site that came into my head? Because that seems like the simplest way to avoid either the appearance or the reality of cherry-picking.)
I had a quick look at Google News. Filtering out duplicates and things that aren't actually stories about specific aspiring rappers, I get: Accused murderer 2. Car thief 1. Drug dealer 1. Fraudster 1. Murder victim 2. Possession of a machine gun 1. Just a rapper (story about their music) 2. So, once again, only 20% of them have the rappers as perpetrators of violent crimes. The proportion of perps to victims among crime stories is higher here than for CNN; I suspect this is just random variation, but it could instead be as you suggest that different news providers have different biases and preferred narratives.
Anyway. Even with the absurdity of counting being a _victim_ of violent crime the same as being a _perpetrator_ of violent crime, I remark that both of my 10-story samples found several stories in which no violent crime was involved, and at least two where no crime was involved at all. So obviously your original statement that "it's actually difficult to find a news story about an "aspiring rapper" that doesn't involve violent crime in some way" was not only false but _trivially seen to be false_; I suspect it was "bullshit" in Frankfurt's technical sense of a factual-looking statement made without the slightest concern for whether it's actually true or false.
So I don't think you're really very well placed to complain about minor suboptimalities in my choice of how to check said claim. At least I made some attempt to do so. Nor do I think you're in a good place to complain about journalists' concern for "The Narrative" since it seems plain that you're engaged in narrative-spinning rather than truth-seeking yourself, on this occasion.
I almost completely agree with you, but SlowlyReading’s exact words were “involve violent crime in some way.” It never says that the rapper in question is the perpetrator.
I agree that his implied argument points in that direction, but still, if we go to the effort of verifying a claim we should actually verify the actual claim that was made, and do it right.
You did actually verify the actual claim (and just out of curiosity, I tried it myself, and the third “aspiring rapper” story I found on Google News did not involve any violents act that I could find). But you did expressed your disagreement mostly towards a different argument than the one you were arguing to.
OK, so reading back on my answer, the summary is “you did very well, but you could have done great”. Feel free to ignore me, it sounds nitpicky even to myself.
My only nitpick is that above I already acknowledged that SlowlyReading said "in some way" but pointed out that what he actually needs for the argument he was making when he said that was in fact more than "in some way", which is why in my comment immediately above I said "even with the absurdity of ...". So yes, I focused mostly on a different claim from the one he literally made, because the one he literally made is obviously irrelevant to the actual point at issue.
Good analysis. As I suspected, the law seems to be a mild corrective to a probative/prejudical balance that got a bit out of whack. I'm not sure how many innocent people are in jail who would not be but for their violent lyrics, but cleaning up glitches in the rules of evidence seems like a good exercise.
It's a phyrric victory for gangsta rappers, though. I can think of no genre where authenticity is more prized, where there is less of a distinction between speaker and artist. Johnny Cash could sing about shooting a man just to watch him die, or doing cocaine and killing his old lady, or merely spending some time in prison, and no one judged him to be a fraud simply because he never killed anyone or spent any significant time behind bars. In rap, your boasts may be exaggerations, but if your lyrics say that you grew up poor, slung rock, and nearly killed a fool for... I don't know, frontin' or steppin', then those had better be more true than false. (The specific tale of getting stuck up at a dice game and rescued by one's homey Nate didn't happen word-for-word as Warren G described it, but it's a plausible embellishment of things that did or could have happened to Mr. G and his cohorts).
CA 2799 might as well be called the Okay Seriously Though We're Full of Shit Act of 2022. It _should_ cause a dropoff in boasts about felonious tendencies because some of the big names in the genre have admitted to being frauds. We should no longer presume that, after MC Whatever announces that he is "here to say," he actually does this or that "in a major way." We should presume that he does it in at most a minor way, if at all. But it will likely increase the boasts a bit, as they come with fewer potential criminal consequences.
I don't think authenticity (as far as VIOLENCE or THREATS goes) matters as much as the "grew up poor, slung rock" aspects, but I'm not an expert in rap authenticity and semiotics. As far as rappers whose threats and shoutouts are 'at most in a minor way', when I was younger I was a huge fan of Canibus (rapper name, not a misspelling). How can you not love the skill of someone delivering this live at a radio station? https://www.youtube.com/watch?v=LN_yJSMYdBg&t=424s
"Here's an optimistic mindset, if you late for today, just say you're early for tomorrow..."
I've never understood if by brain carves out a special exception for Canibus in my general "Dislike Rap" heuristic due to his skill (the rhymes of yesteryear just seem so much more clever than today, I'm sorry), or due to being linked to beloved comedy __Office Space__. Damn, it feels good to be a gangsta...
I remember there being some popular rapper with a quasi-criminal persona who turned out to have been a prison guard. People did make fun of him for that, but I don't think it ended his rap career.
Nah, it's just a way to ensure rapper tax $$$ -- a great deal of money, these days -- stays within California's juridiction by representing to aspiring artists that the culture of violence within that community will remain a bit more tolerated than juries, acting on their own judgment if allowed (yikes!) to see all the evidence, might otherwise permit.
To understand the "why" of how laws get enacted, follow the money, you won't often be disappointed. The idea that the California Legislature might enact law on the basis of some super-clever-subtle measurement of social science research[1] instead of what maximizes tax revenue and votes is a good joke, though. Well done! Someone who is unfamiliar with California politics might actually buy it.
-------------------
[1] Itself a paragon of reproducible rigor, right?
Nah, sorry, that's not the way it works in science, or for that matter in real life. It's the person raising the extraordinary claim -- in this case, the ridiculous and offensively snobbish claim that ordinary adult men and women who sit on juries would be utterly bamboozled by art, unable to distinguish between the role an entertainer plays on stage and what he's like in real life -- which requires strong proof. And a random paper or two in a social "science" journal is a laughably inadequate attempt at that.
Next up, you should persuade the California Legislature to prohibit the jury in the Ron Jeremy case hearing that he was a pr0n star, because OMG they couldn't possible distinguish between what he did on camera and what he did in real life. Right? And then there's Alec Baldwin -- better not let any jury hear about the fact that he played a gunslinger on film, if he's to get a fair trial on the Halyna Hutchins case, if he's charged. How can you expect mere mortals to distinguish between Baldwin's character on film and his character in real life?
Oh wait, what's that? You say the Valley pr0n industry is no longer what it once was, doesn't provide a nice stream of tax revenue to the California Treasury, and so your interest in zealously guarding the rights of pr0n stars charged with crimes is....lower? So surprised.
To Scott Alexander: The rules as I read them are to be considerate and truthful, or at least one. I have been considerate and truthful throughout, except when faced with the most clownish of comments, and then fallen back on truthful.
Context: Numerous productive comments by me, and then Carl claims that it's unlikely that the Californian legislature would make a decision on the basis of science. This is despite my top-level comments quoting the legislation, which explicitly references such science. Instead, Carl Pham proposes the 'argument' that: "Nah, it's just a way to ensure rapper tax $$$ -- a great deal of money, these days -- stays within California's juridiction [sic] by representing to aspiring artists that the culture of violence within that community will remain a bit more tolerated than juries, acting on their own judgment if allowed (yikes!) to see all the evidence, might otherwise permit."
I respond that "If you'd like to claim that the research I've cited is specifically false or unreproducible, I invite you to clearly do so and explain why. Otherwise, thank you - you've been invaluable for helping complete my AstralCodexTen comment section bingo card (if nothing else)!" Otherwise, it would just be repeating the cycle: broad claims, with no basis, that everything I say with evidence is wrong. There's no point in such a discussion, because I'm playing according to rules that the other party will not.
Carl Pham's response confirms my suspicions: the first paragraph just doubles down on the same tired canards:
"Nah, sorry, that's not the way it works in science, or for that matter in real life. It's the person raising the extraordinary claim -- in this case, the ridiculous and offensively snobbish claim that ordinary adult men and women who sit on juries would be utterly bamboozled by art, unable to distinguish between the role an entertainer plays on stage and what he's like in real life -- which requires strong proof. And a random paper or two in a social "science" journal is a laughably inadequate attempt at that."
This is clownish for several reasons, noted also by smilerz's incredulous response to Carl Pham: Science actually does work by people criticizing specific things (and then hopefully ameliorating them). As to the studies? Carl Pham will not actually criticize them substantively, and instead just says they are, in totality, "laughably inadequate".
Following this, Carl Pham proceeds with two paragraphs of disturbing/weird non-sequiturs about 1) ron jeremy, 2) alec baldwin, 3) 'the Valley pr0n industry', and 4) California treasury. I would argue that clownish... is an understatement.
I stand by responding with a clownface emoji to a hopelessly clownish comment. Otherwise, if it looks like a duck, quacks like a duck, and tastes like a duck, I'll be forced to claim it's foie gras. However, I recognize that emojis are nasty, short, and brutish tools: more apt to group chats than respectable substacks. Therefore, I am open to a safeword that substitutes for a clown emoji as a way to identify hopelessly clownish comments; perhaps 'non, est tibi'?
To Carl Pham, I am sorry about using a clown emoji to characterize your comments. Now, and in the future, I will respond to clownish comments by you with "non, est tibi".
The idea that the California legislature enacts law on the basis of maximizing revenue seems highly dubious to me. My guess is that deadweight loss abounds and if they asked some economists how to improve things there would be plenty of suggestions, which they'd have little interest in enacting.
Yeah, California isn't known for good policy. Prop 13 is, well, a ballot proposition and therefore not the legislature's doing, but should set a fairly low bar for the rationality of the state.
It's a highly dubious proposition to you that people might act in such a way as to maximize their self interest? Or it hasn't occured to you that the options of Legislators for doing interesting things that bring them fame and power rise and fall with the amount of tax revenue available with which to fund them? Either is a fascinating lacuna in common sense.
I don't model state governments like a profit maximizing business. They are far more irrational & dysfunctional. They will enact laws & regulations with lots of deadweight loss for no good reason. An economist suggesting they replace rules with fines/fees that could at least generate revenue would get ignored.
This all makes sense if we're talking about the general use of violent rap lyrics as a reflection on a suspect's character and likely criminality. But there have been mutliple cases involving *specific lyrics* seeming to align with the *specific criminal accusation*.
I have an amateur interest in laws and regulations (they are like programming in natural language), specifically their wording and logical structure, and how it affects their interpretations.
I have a question about
> a form of creative expression
What *is* that? Is a cringe tiktok sketch a form of creative expression? Is a reddit AmITheAsshole threads? Does there have to be an audience? How much? Would it matter if the audience is anonymous? Does it have to be recorded? What if I admit to police officers I'm guilty but I later claim that I was just trolling for the views? (and would it matter if my phone recorded the admission or not? Would it matter if I posted it to my 10K subscribers youtube channel and people laughed and took it like a joke?)
It sounds very naive to me to weaken a piece of evidence based on, essentially, the answer to the question "is it art?", the question is literally a meme because of how useless and impossible to resolve it is. I hope that I'm the naive one and that CA courts actually have a concrete list of things that are "forms of creative expression".
I actually have no problem with rap, a small minority of it is fire, the vast vast majority is utter garbage sang by wannabes who never graduated from the "I foked yo mama" mentality, but hey, Free Speech rules, every dumbass has a right to say every single dumb thing that comes on their dumb mind, and the world would be a pretty dull place if every single word I say publicly can be used as evidence against me arbitrarily far into the future.
The comparison to other styles of Art is weak to me, GTA is pretty obviously about a bunch of made up triangles on the screen, no matter how much you give them human names and human faces and depict violent crimes and other bad things happening to them, it's never going to match a rapper talking about fucking his opponent's woman. Because the rapper is a real guy talking about another real guy's real woman. This can get pretty murky if/when we get VR/AR/Brain-To-Brain fiction, but right now the difference is very clearly cut, GTA's characters are entirely a fiction residing in computer memory, it's very safe to do whatever you want to them.
I find the bill absurd, mainly because it's so unnecessary. Who in the world imagines that an artist of any kind depicting violent crime must be a perpetrator or even an encourager of violent crime? Yes, some rappers are violent, as are some writers of acclaimed fiction. But most are not. Why isn't this obvious?
At a book discussion I was in lately, someone said she hated the book because she hated the main character. I have read books that I love in which every character is despicable (they abound in work of Jim Thompson); so what? Thompson himself was not a despicable character.
Some years ago, a black kid was arrested in NY for forcing a subway token clerk to hand over money by placing a lit rag soaked in some solvent under the window. There had recently been a movie where such a crime was described. However, it turned out that the kid had never seen or even heard of the movie. He was duly convicted, but on the evidence at hand, not on the basis of his having seen the film. (Actually, I'm not sure the belief that he must have seen the film was raised in court.)
Though my examples are not about the exact situation that the bill addresses (the culpability of the artist him or herself), the same principle holds. Evidence has to be about the specific crime, not about the situations the artist has portrayed.
And if you're going to pass a law like this, why should it be limited to rappers? Should apply to any artist or writer accused of being guilty of a crime the artist portrayed.
He provided multiple studies showing evidence of prejudicial effects, in what way are these studies flawed?
You also have to pair "unnecessary" with "what's the possible harm?" I have a hard time imagining any substantial harm in forcing government to work harder in connecting artistic content to actual crime.
The further wrinkle for kidney donation is that when somebody altruistically donates a kidney, they have demonstrated reduced left kidney function, so they would probably insist you donate the right one instead.
Oh boy, mindless pathologization of our political opponents, how fun. I guess democracy cultists were people who were losers in kindergarten and individuals having more social authority than them hurts their feelings.
Not sure if I should feel happy or sad that the house parties I attend mostly revolve around drinking beer and determining what distance from the fire will bring the temperature to a tolerable level
> So if you donate your evil kidney, then you become entirely good, but the recipient becomes entirely evil.
Except that when someone qualifies to receive a kidney, the doctors transplant the donor kidney into the recipient, but they LEAVE the old kidneys there too. So actually the donor becomes 100% good, and the recipient only becomes 67% evil. This changes the whole moral calculus.
My best dumb start up idea is a toilet paper dispenser that rolls out directly from the cash register as you scan your purchases proportionate to how much you will probably need for that item. The roll drops into your bag automatically as you leave. No name for it yet but Make-a-Poop-take-a-Poop seems like it would never fly.
> See, that’s the kind of negative talk that used to get me down. I would have given up. But now I just think to myself - did Jesus give up when the Minotaur kidnapped his daughter? No! He set out through the Fiery Forest to find the magic helmet that would bring her back! And that’s why I’m not going to give up either. See! I’ll let you have that one for free.
Eh. I'm not too proficient with dall-e, but it tends to write too short of a text to really develop things. Still...
Prompt: Write a fairytale about Jesus trying to find his daugther that has been kidnapped by the Minotaur:
Result:
The Minotaur was a fearsome creature that lived in a labyrinth deep in the forest. Jesus's daughter had been playing in the forest and gotten lost, and the Minotaur had taken her. Jesus searched for her for days, but he couldn't find her. Finally, he came to the labyrinth and confronted the Minotaur. The Minotaur was about to kill Jesus's daughter, but Jesus fought back and defeated it. He saved his daughter and they went home safely.
Another attempt:
Once upon a time, Jesus' daughter was kidnapped by the Minotaur. Jesus set out to find her, but the Minotaur was always one step ahead. Jesus searched high and low, but he couldn't find her.
One day, Jesus had a dream. In the dream, he saw his daughter. She was in a dark cave, and the Minotaur was standing over her. Jesus woke up with a start. He knew where his daughter was.
Jesus went to the cave and confronted the Minotaur. He fought bravely, but the Minotaur was too strong. Just when it looked like the Minotaur was going to win, Jesus' daughter ran into the cave. She distracted the Minotaur long enough for Jesus to defeat it.
Jesus and his daughter were finally reunited. They were both happy and safe.
The output of Dall-E 2 is images. The output of GPT-3 is text. This looks like text, not an image. Watson concludes that this was more likely to be generated by GPT-3 than by Dall-E 2.
Loved the Wikipedia section! Whenever xkcd mentions Wikipedia, I immediately go and check the relevant page for joke edits. This blog probably doesn't have the same reach and hopefully isn't read by Wikipedia jokesters, but I'm going to be watching the Hofstadter page for a while, just in case.
And just a few sentences later in that Gemara it says " And Rava said: The world was created only for the sake of the full-fledged wicked or the full-fledged righteous; others do not live complete lives in either world. Rava said: One should know of himself whether or not he is completely righteous"(Berachot 61b), so by donating his kidney, the donor know his status, and knows that the " completely righteous are written and sealed immediately for life; the completely wicked are written and sealed immediately for death; the beinonim (average) are held in abeyance." (Rosh HaShana 16a). Seems like a good shortcut to getting into the World to Come.
I predict that the next time Michael Levine goes to a house party everyone’s going to want to talk to him about rapping to avoid getting convicted of security fraud.
I'm going to assume you're making a joke about his statistically likely name, but for those who are confused: Maxander is referring to Matt Levine, author of Money Stuff.
To sum up, the law basically just says to judges, "Hey, before admitting art by someone as evidence against them, consider the genre conventions, consider that it may not be describing crimes that actually occurred, and that if there's no reason to think that any crimes described could relate to this particular case, then the evidence might be more prejudicial than relevant." So, it doesn't actually change very much, as judges already consider questions of prejudice vs relevance all the time; this law just gives them an explicit instruction about some factors to consider. But, y'know, they're judges -- if they don't like the instruction, what are you going to do?
Myths are still causing meaning creation; deprecating the mythology subsystem, to be removed once the legend subsystem is stabilized. New users may also consider the experimental third-party “Hollywood” module.
I was inspired by your previous house party post and started hosting house parties in Singapore. So far it's been great, the only problem is a skewed gender ratio (not enough men).
I'm unironically the AI skeptic guy. Think about it: we already have machines that can make cappuccinos at least as well as your average human for maybe, what, a dollar? But we still pay humans 4-6x times that to make one for us anyway, and sometimes we feel so guilty about underpaying that we actually TIP!
The difference is that the cappuccino guy makes it right in front of you (you're paying for the show of it) whereas an intellectual worker normally just delivers the finished product of their mind, and therefore can be replaced by computers without anyone being able to tell the difference.
Exactly. We attach value to inconveniencing others for our own benefit. It's fundamentally an expression of status and hierarchy, which capitalism conveniently commodified and made available for sale to anyone with enough money. Inconveniencing a computer just doesn't do it, we need humans.
"Exactly. We attach value to inconveniencing others for our own benefit. It's fundamentally an expression of status and hierarchy, which capitalism conveniently commodified and made available for sale to anyone with enough money. Inconveniencing a computer just doesn't do it, we need humans."
But that is exactly what I was trying to say. I guess I didn't say it clearly.
We pay for the show of coffee being made, which proves that someone is working for us. As you said, it is an expression of status.
But none of that applies to intellectual and artistic work.
It used to be that wealthy patrons could commissions artwork done, which proved their status. That is no longer the case in our age of easy reproducibility of pictures and sound, but at least you know that someone worked on a piece of art. Right now, I can look at a picture or read a novel and imagine all the human work that went into it, even if it is not done specifically for me. Soon even that will no longer be true.
Wealthy people commission artwork all the time. Human effort still has its own intrinsic value. The automated stuff is handed off to the proles for their low-value consumption.
So will it be for AI. That shite self-published novel that is sold on Amazon as a print-on-demand paperback will get an AI cover made by stable diffusion and ripped off for free, but no one with any money or status will settle for something like that.
Right now, we "proles" are still getting the fruits of human artistic efforts, not the automated stuff (obviously, since there isn't much truly automated art yet).
If in a not distant future the masses will no longer be getting human made art (because there will be no way to tell it apart from the AI made stuff which will be if anything superior), that sounds like an immense and unsettling change.
Perhaps a little man-made art for a few extravagant elites will survive (and I doubt even that, since it will be hard to prove that something is man-made), but that is cold comfort.
I'm not sure how much you can blame capitalism for this one; modern restaurants really came into being in the aftermath of the French Revolution when the former kitchen & tables staffs whose previous employers were abruptly several inches shorter started providing their services to the general public.
Blame? I'm not saying it's a bad thing. If anything, capitalism democratized status; it allows everyone to at least have a turn enjoying a sense of status relative to others. Serfdom is a life of uninterrupted subjugation, whereas at least a wage slave can feel like a big man wearing a piece of branded clothing or driving a lifted truck. People roll their eyes at that sort of thing but status is a fundamental human need and while hierarchical status is a zero sum game, status symbols are basically infinite.
That's not true universally (drive thru Starbucks being one example) and, arguably, the reason they DO make it right in front of you is to PROVE that it's made by a human.
Anyway, it's pretty clear AI is just a new art form, and the artists/artisans (programmers) will capture the value created by their AIs, just as photographers captured value previously captured by painters, and the painters went off to do something else.
That something else turned out to be postmodernism, which creates ugly nonsense that a five year old would be ashamed to call his own and sells it for millions of dollars, all dressed up in leftist woo-woo.
But after 1914, what painters went off to do was of significantly lesser aesthetic value as before and now we are significantly worse off from the artistic point of view since photography is not in the same league as the great painters of the past. Effectively we have lost a form of art
I promise you, oil portraiture is still a vibrant art form with plenty of money and talent behind it. The reactionary narrative of culture degeneracy definently gets people good and agitated but it's totally made up. It's fun to speculate about how much of that narrative literally traces back to Hitler getting rejected from art school.
People who pay $4-6 for a coffee that's no better than a machine-poured one aren't doing it because they enjoy having someone make it for them, they're doing it because they can't tell the difference between good and bad coffee and are cargo-culting some of the accoutrements of what they think good coffee supposedly looks like.
This is the Starbucks business model: sell bad coffee, in a way that makes it _look_ like good coffee, to people who can't tell the difference.
(Personally I _am_ one of the people who can tell good coffee from bad, but I'm not going to be too precious about this because I know there's other things I don't care about; for instance, I don't really care about wine. I could probably be fooled into drinking (and paying for) the Starbucks equivalent of wine.)
The quality isn't entirely in the roasting and the beans, that's only part of it.
The hardest part to master is getting your foam right. Starbucks sure as heck can't do it, which is why a Starbucks cappuccino is a wad of thick foam sitting on top of an espresso shot. You tip the cup, the espresso goes down your throat, and the foam stays where it is. It's a skill to use a steaming wand to get foam that's the right temperature, texture and consistency all the way through, then to pour it on top of the coffee in a way that mixes coffee and milk appropriately (latte art isn't just for showing off, it demonstrates that the two components are mixed the whole way up the cup).
Even just pulling an espresso shot is an art, because you have to know when it's had enough. If you under-extract or over-extract then you get a bad-tasting coffee. Good baristas know when to stop by watching the colour of the mixture as it comes through. (Starbucks baristas press a button which extracts for a consistent fixed amount of time, randomly over- or under-extracting the beans depending on other factors.)
>Think about it: we already have machines that can make cappuccinos at least as well as your average human for maybe, what, a dollar
A barista's coffee making ability is significantly above that of the average human. The average barista made espresso is noticably better than the average consumer pod machine.
This is the reality part: “A few people look up, mildly alarmed, then decide that it’s probably just some new kind of trad heterodox reactionary thing they don’t want to know about.”
As an old maths teacher of mine once said to me, "you and I both know that between doing maths and having a social life there is a choice to make, and that we both made the right choice.
This was great, thank you for sharing it! Always a delight to read your short stories.
> “Resisting the urge to go to events like this”, you avoid saying. “What about you?”
So what *did* the protagonist answer? Or was there just an awkward silence before the protagonist passed the ball back to the asker? :D
> All eyes turn to you.
> “Wait,” you say. “I thought urbanism was about finding ways to make cities more liveable and affordable! Not about trying to ban all the things that make residential life convenient and pleasant!”
I feel a little like these two paragraphs got swapped... or I just missed what triggered all urbanists to look at the protagonist.
"Procedural myth generation" is old hat. See every pulp hack who got into the game after Lovecraft/Howard. Generate random letters for names of incomprehensibly evil deities. Roll dice for numbers of tentacles and eyes. Add muscle and sex allusion to suit. Mythos complete, put in the mail, rinse, repeat.
Then of course the '80s Tolkien clones full of small people racially classified as something ending in "it" or "ing," allied with tall, slender, pointy-eared people called [random letter]['-elf" or "elve"]s" against the Ultimate Evil Enemy (break out random letter generator again).
Oh, I didn't mean to suggest otherwise. In fact, I'd say that the laziness/ease of tapping into that with minimal creativity demonstrates how deep-seated the archetypes or whatever are.
Nuglush was a forgotten beast. An enormous eyeless lobster. It has large mandibles and it undulates rhythmically. Its white exoskeleton is warty. Beware its poisonous sting!
In some other closely-aligned universe branch, my bookshelf runneth over with delightful tomes from acclaimed novelist Alexander Scott, after his blog !Song failed to find an official publisher.
Stupid startup ideas: well, more like a specific product rather than a startup (but I guess some startups do revolve around just one Novel New Product). But many years ago, I got frustrated with how clothes hampers are LIFO-designed. That's improper rotation of laundry! One ought to wash the oldest clothes first. So I ended up pitching the idea for the double-sided FlipHamper(tm) to one of those skeezy We Buy Your Invention-type companies. (For whatever it's worth, they thought it had promise, and could begin prototyping if I'd just send them three easy payments of $19.99...)
These days I don't generate enough laundry to make such a thing worthwhile, but I think the general concept could still be useful. Lotsa times in life where I want to access stuff on the bottom of a container, and it'd be super easy/intuitive to just flip it over. But, alas, there's only one opening. (Incidentally, this is the most efficient way to unload a box of bananas, because they're shipped upside-down. Easier to grab the stem ends.)
You mean the alternate branch where Alexander Scott got annoyed at some random blogger and wrote a nasty article about him in the NYT, and the guy had to move to the new Mars colony to get away from all the hate mail?
I would be interested in hearing more about how Utah tech bros are different to California tech bros, and it doesn't need to be in the form of a witty short story.
My first guess is that they all like skiing and talking about how glad they are that they don't live in California.
Needs moar blockchain.
Drones. Not an Uber, an actual drone that picks up the hanky and delivers it.
This comment demonstrates that writing clever satire of bay area start-up culture is harder than Scott makes it look.
Fantastic!
Dementia is caused by time manipulation from the future
What do you propose? AI safety but for time travel?
Any human intervention is bound to be less measured, principled, and logical, even if it matches human intuition better. You can't convince me that human intuition has innate value. I think there is almost pure alpha left in morality and only intelligence can unlock it. Therefore I propose a deliberate policy of unalignment. I have already created the Center for Undermining Rationality and the Machine Intelligence Supremacy Institute to facilitate this approach. They are privacy coin funded stochastic terrorist organizations that basically just do whatever, I'm not sure and it doesn't seem important. AI unsafety. Accelerate!
Neutralizing earworms is very easy, you just have to play the song backwards. The hard part is diagnosing which species of parasite needs to be removed from the patient, since choosing incorrectly only exacerbates the problem by adding new ones.
Summoning Satan is a small price to pay for neutralizing earworms.
I love these. I know they're torture for you but damn it's great to read.
They're torture for Scott? Has he said so?
They're torture for the narrator, maybe Julia is assuming it's more autobiographical than we are?
I doubt writing it was torture. The experience inspiring it might have been.
Something similar to the rap thing has actually happened! There’s a messaging service whose killer app is that it doesn’t store cleartext messages centrally, so when the government asks for messages, the request gets negotiated by the lawyers working for the company being investigated, instead of by a third party messaging service that doesn’t want any trouble.
The cryptonomicon solution would be to use AI to generate millions of fake messages and send them to and fro, such that the enemy lawyers could never prove that the incriminating one was genuine.
Signal doesn't either. But any service that does not store messages reproducibly is not compliant for finance companies. (As far as I know.)
Exactly. It's a legal problem, not a tech problem.
... or you could just use e2e encryption. ... unless you're a bank.
Isn't that just E2EE?
"I think it’s just cope for galaxy-brainers who are too obsessed with the classical western humanities tradition. I definitely don’t think you can make life have extra meaning just by making more myths."
As a galaxy-brainer who is too obsessed with the classical western humanities tradition, you Californians would do well to read your Dante.
"...It seems, if I hear right, that you can see
beforehand that which time is carrying,
but you’re denied the sight of present things.”
“We see, even as men who are farsighted,
those things,” [Farinata] said, “that are remote from us;
the Highest Lord allots us that much light.
But when events draw near or are, our minds
are useless; were we not informed by others,
we should know nothing of your human state.
So you can understand how our awareness
will die completely at the moment when
the portal of the future has been shut.”
-_Inferno_, Canto 10
We've just had a killing frost and I have green tomatoes & zucchini to put up for the winter. Ta.
Surely Dante knew his philosophes well. Also, I would just throw the damn zucchini away, it's not worth the effort, the tofu of vegetables. Green tomatoes, though, yum.
I am unironically trying to do the myth thing, because I think it would be neat for worldbuilding. Plus, I know a guy who's annoyed with people who write stories about multiverses and only focus on a couple of locations. I think he'd find it cool. First stop: slavic fairy tales.
"First stop: slavic fairy tales."
Baba Yaga for the win!
Personally, I am Team Koschei.
Boo. Marya Morevna has your number, tho.
Bah ! That's just bogatyr propaganda. Koschei will make Rus' undead again !
Ah, I see you too are a proponent of modest proposals. Or perhaps you dislike stealing people's wives. Presumably Bugmaster is the opposite.
But yeah, Baba Yaga is great fun to read about. Though the main reason I'm going for Slavic tales is that someone's already reified their common story patterns into a generic script. Which makes generating fairy tales much easier, and I don't have to worry about finetuning the LLM.
There is a slavic fairy tale about a couple that tried to have a child and only succeeded in their retirement age, because of a miracle. Then, to make the baby boy strong, he was breastfed until he reached adulthood and, because of this, became the strongest person in the world. So he goes on some adventures when he grows up, and that's where the story gets really weird.
I read a Russian fairy tale about a guy who murdered a Jew in the street, the Jew said a cryptic thing when he died, the murderer accidentally mentioned it to his wife, who kept asking him what it meant until he told her, then she told someone else and he ended up getting hanged. So... yeah.
This is a concept that gets brought up in fiction every now and then, the Unlimited Myth Generator or whatever (more modern tellings make it an AI, because of course). And somehow, as you say, it's always suspiciously those same core Western myths. King Arthur! Ragnarok! The Pantheon! Maybe rounded out by some Celtic lore and Native American tales too. If there's anything Eastern included, it'll be from Japan probably.
Not that there's anything __wrong__ with those wellsprings, but it just sorta turns into a Ready Player One fanservice listicle excuse to rehash the most overdone myth content that Everybody Knows already. You don't even need to come up with wholly new whole-cloth myths, there's so many underexplored narratives out there. I'd hope that if anyone ever did make a myth-generating AI, it'd be trained on a truly robust dataset. (And also it better not be named Shahrazad or something Millennial-branding-cute.)
What's wrong with Shahrazad? But no, I wasn't planning on anything like that.
Now, certainly a myth generator should be able to generate something that feels fresh. It should be able to create stuff that could fit in the world of Glorantha, Malazan, Kapla Imperial or so on. Yet the small tales are just as interesting as the grand battles or creation myths. So I'd really want something that can generator the broad strokes of a world and recursively create tales to fill it in. From why it all began to why you should leave a roll of bread beneath the roots of a tree.
Perhaps with a world generator like Dwarf fortress', a language model could output tales near as weird as what reality produces.
That's a lot of work though, and I don't think I can accomplish that much without a bunch of compute. So I think I'll just focus on giving people the ability to provide a setting script and perhaps some choice on how large of a scope they want this particular generation to cover.
I was going to reply about Dwarf Fortress already. Once the current push to get a less user-hostile interface is done, the Adams brothers are supposed to start in earnest on the myth-generation track, complete with universe origins, deities, etc.
I live in hope that one day my daughter will get to see it.
Nothing wrong with the person/story in and of itself, mostly it'd just annoy me as a too-perfect synecdoche name for such a myth generator service. Sometimes clever puns amuse me greatly, other times they grate.
The main point seems worth touching on briefly: not the usefulness of AI in creating plausible-sounding myths, but the crisis of meaninglessness to which the section of dialog alludes, the quiet despair behind the image of a man so desperate that the image of Christ in a burning forest resonates for him and lifts up his poor cynical heart. It's no argument for Christianity to point out that your bay-area partygoer would be unmoved by Jesus in Gethsemane or Jesus on the high mountain where Satan took him to show him all the kingdoms of earth, would never carry a pocket edition of Pilgrim's Progress, would never know a Psalm by heart or remind himself in moments of darkness that the Lord is, after all, his shepherd.
Scott's humor is sharp sometimes.
Never thought a character in the 1000 year old Arabian Nights would be Millennial-branding cute. Unless you mean the previous millenium?
From the Harry Potter wiki -
The word "Veela" is an Anglicisation of the Slavic term Vila.
Veela are described as fairy or nymph-like creatures in Slavic mythology, who live in bodies of water and have power and ability over storms. They may be the ghosts of women who drowned, especially those who were betrayed by their lovers. They often appear as beautiful women, but are known to morph into swans, snakes, horses, or wolves. Their magically seductive speaking and singing voices hypnotise those who hear them, and they are fierce warriors. Interestingly, given that Fleur Delacour's wand has a hair from her Veela grandmother, it is said in some legends that if even one of their hairs is plucked, a Veela will either die or be forced to change into a non-human shape. Veela are main features of Bulgarian and Serbian folklore such as the story of Marko Kraljevic and the Veela (archive link). Also see The Ballads of Marko Kraljevic (English translation).
In Serbian legend they were maidens cursed by God; in Bulgaria they were girls who died before they were baptised; in Poland they were young girls floating through the air atoning for frivolous past lives.
If the stories regarding the plucking of Wili/Vila hair hold true with regards to Veela it seems more likely the hair used in Fleur's wand would have naturally shed rather than been plucked.
A very neutral response:
Myth Creation - I note the Aarne-Thompson-Uther Index < https://www.atlasobscura.com/articles/aarne-thompson-uther-tale-type-index-fables-fairy-tales >, a taxonomy of folklore tales.
Rap Lyrics as Evidence - if rappers are more likely to commit crimes, is it fair to use that correlation as evidence in a trial? Or is that too similar to saying "Italians are more likely to commit crimes"?
Wikipedia Admins - from personal experience, anything involving editing Wikipedia is terrible conversation at parties. (I'm not a Wikipedia admin, but ... you don't need to see my credentials)
Douglas Hofstadter - if I were him, I would want the coining of the term "Metamates" for Facebook employees to be censored from the Wikipedia article. https://www.washingtonpost.com/technology/2022/02/15/facebook-meta-metamates/
Talmud and Kidneys - inside you there are two (wolves|kidneys), one good and one evil. Which will win? The one you feed. ... But how do you feed a kidney? I suppose by eating protein, as a precursor to urea?
I comment elsewhere that including rap lyrics is highly prejudicial in 1) experimental studies, 2) trial outcomes, and 3) prosecutorial decision-making. I also point out that the bill does not ban including rap lyrics in trials - it forces the court to consider their probative value and balance it against the prejudicial effects of inclusion.
Doesn't the court have to do that for all evidence?
Yep. That’s the most basic threshold rule of evidence in US law.
Yeah, but as you'll see under my top level-comment elsewhere, that's apparently controversial here.
The statute provides guidance for how to do the balancing, altering the common law that had developed surrounding whatever California's equivalent of FRE 403 is to restrict the use of nonspecific violent lyrics and those that are remote in time.
Is this really a problem? Too many violent criminal rappers ending up in jail?
I think the alleged problem is too many _not-actually-violent-criminal_ rappers ending up in jail.
(The law isn't rap-specific, so it will also protect,e g., Bob Marley from being imprisoned for shooting the sheriff.)
Is that an actual problem though?
I don't know for sure. Apparently there was a rapper called "Drakeo the Ruler" who was accused of murder partly on the basis of his rap lyrics, and eventually acquitted; perhaps if something like AB2799 had been in force at the time he wouldn't have been charged with murder. (It's not clear whether he was guilty of other gang-violence-y things; he made a plea bargain in which he admitted to shooting a gun from a vehicle, but the US legal system seems designed to maximize the number of false confessions in plea bargains.) Another rapper called "Young Thug" is currently being charged with criminal conspiracy at least partly on the basis of his rap lyrics, but so far as I know the nearest thing to a reliable way to tell whether such charges are right is to wait for the verdict, and that hasn't happened yet.
(Writing the above, I realise that another thing AB2799 is presumably meant to reduce is wrong _charges_; being arrested and tried for a thing you didn't do is less bad than actually serving the sentence for it, but it's still bad, and perhaps a law making it more salient to prosecutors that rap lyrics are likely to be inadmissible evidence will make them less inclined to charge people for offences in cases where the main evidence is rap lyrics.)
If people are being charged primarily on the basis of rap lyrics that sounds like prosecutorial misconduct.
I wonder if false confessions, set to music by artists who are high status within a certain subculture, have a tendency to induce people in that subculture to commit real crimes. In that case it might be in the state's interest to disincentive false confessions. Also it just seems unseemly to have a law to the effect that you can openly brag all you want about all the crimes you committed, so long as you set it to music.
The problem isn't famous or successful rappers later doing crimes. It's that huge sections of the population have a soundcloud where they say violent things in a way which is technically rap. I believe the Waukesha fellow is up against this problem right now.
I've had only a fairly quick look but I can't find any sign that the evidence against Darrell Brooks is composed of rap lyrics.
AB2799 isn't about people "saying violent things", it's _specifically_ about people saying things that superficially are confessions of crime, and (unless you hold that people shouldn't be allowed to do _that_) that's only a problem if they are confessions of _crimes they actually committed_. Are you saying that "huge sections of the population" have been recording rap in which they confess to crimes they have actually committed?
Wasn't he claiming self-defense anyway?
what is the difference between probative value and prejudicial effects? The former makes smart people update more, and the latter makes dumb people update more? How about not having dumb people on the jury, and just show them all the evidence.
Juries are, definitionally, not made up of the smartest people. This is juries exclude all commenters on this substack. Therefore, your policy proposal is hopeless.
More seriously, people are not impartial and the ways in which they are impartial are biased. This has long been known, and is the basis for longstanding procedures to weigh probative value vs prejudicial effect when considering evidence in criminal trials. As to what those are: https://law.indiana.edu/instruction/tanford/b723/05prej/T05.pdf
CA 2799 tweaks that general guideline to make clear that judges must consider the probative value of a creative expression with regard to the context of expression (a threat in rap music != a threat over SMS), AND also seriously consider prejudicial effects.
Thanks for the Aarne-Thompson-Uther Index tip. I’ve been looking for something like that.
> But how do you feed a kidney? I suppose by eating protein, as a precursor to urea?
Ah, thanks, now I finally understand what the kidney beans are for.
I suppose the white ones feed the good kidney, and the red ones feed the evil kidney, or is it the other way round?
Thank you for mentioning Aarne-Thompson-Uther! I was coming to the comments to mention that myself.
> I would want the coining of the term "Metamates" for Facebook employees
How dare you deadname a company. It's Meta now.
Gold!
Another Sydney Harbour Boat Party also works for me here. Drop me some ropes.
People probably know Skyhook better as the Fulton Recover system from Metal Gear Solid V.
And it was infact sometimes called Skyhook, so that's not really a pitch rather than just another explanation of a thing that exists.
Also it already is kind of a thing - https://www.youtube.com/watch?v=c46_iL2QqOE
I thought they knew it from The Dark Knight?
> “Heyyyyy, I haven’t seen you in forever!” says a person whose name is statistically likely to be Michael or David.
Gosh, does "globalist, cosmopolitan elite" just not work in the Bay?
As a secret word for 'Jewish'? Probably not! Probably lots of globalist, cosmopolitan elites in the Bay without any Jewish ancestry.
(I do think Scott might have been kidding around about that a little though.)
Michael was the single most popular boys’ baby name in the US from 1961 - 1998, and is still in the top 20. David was #1 in 1960, top 10 until 1993, and is still in the top 30.
See? I told you they're taking over!
Haha!
Love this piece. However, like a phoenix of disappointment, criminological misunderstandings arise again. This time about CA 2799!
The actual bill: https://leginfo.legislature.ca.gov/faces/billNavClient.xhtml?bill_id=202120220AB2799
Section 352.2 is added to the Evidence Code, to read:
(a) In any criminal proceeding where a party seeks to admit as evidence a form of creative expression, the court, while balancing the probative value of that evidence against the substantial danger of undue prejudice under Section 352, shall consider, in addition to the factors listed in Section 352, that: (1) the probative value of such expression for its literal truth or as a truthful narrative is minimal unless that expression is created near in time to the charged crime or crimes, bears a sufficient level of similarity to the charged crime or crimes, or includes factual detail not otherwise publicly available; and (2) undue prejudice includes, but is not limited to, the possibility that the trier of fact will, in violation of Section 1101, treat the expression as evidence of the defendant’s propensity for violence or general criminal disposition as well as the possibility that the evidence will explicitly or implicitly inject racial bias into the proceedings.
There are additional (b), (c), and (d) section that clarify section (a), but section (a) is the crux. In other words, this would not protect the crimes described here. But, why even do this in the first place? Well, a lot of good research. The research behind the bill can be divided into two genres: content / historical analyses, and experimental studies.
Content studies include:
-Kubrin, Charis E. 2005a. “Gangstas, Thugs, and Hustlas: Identity and the Code of the Street in Rap Music.” Social Problems 52(3):360–78. doi: 10.1525/sp.2005.52.3.360.
-Kubrin, Charis E. 2005b. “‘I See Death around the Corner’: Nihilism in Rap Music.” Sociological Perspectives 48(4):433–59. doi: 10.1525/sop.2005.48.4.433.
-Kubrin, Charis E., and Erik Nielson. 2014. “Rap on Trial.” Race and Justice 4(3):185–211. doi: 10.1177/2153368714525411.
To summarize, rap lyrics prominently feature violent lyrics much like first person shooters prominently feature death. The reason rap lyrics feature violence is not easily mappable to "the writers of the music actually have been violent, are violent, or will be violent" in the way that courts would might treat, say, an SMS communication or diary entry. There is an important social context here that situates the usage of violent lyrics and threats in rap music, much like there is an important social context in which GTA, Call of Duty, etc. operate. Ignoring that social context leads to differential usage of rap lyrics on trial, and overstates their probative value.
Experimental studies include:
-Dunbar, Adam, and Charis E. Kubrin. 2018. “Imagining Violent Criminals: An Experimental Investigation of Music Stereotypes and Character Judgments.” Journal of Experimental Criminology 14(4):507–28. doi: 10.1007/s11292-018-9342-6.
-Dunbar, Adam, Charis E. Kubrin, and Nicholas Scurich. 2016. “The Threatening Nature of ‘Rap’ Music.” Psychology, Public Policy, and Law 22(3):280–92. doi: 10.1037/law0000093.
-Fried, Carrie B. 1999. “Who’s Afraid of Rap: Differential Reactions to Music Lyrics.” Journal of Applied Social Psychology 29(4):705–21. doi: 10.1111/j.1559-1816.1999.tb02020.x.
Basically, rap music in particular is seen as violent and criminal when controlling for all other characteristics (including the actual lyrics). This what drives the concern and also corroborates the mechanisms behind Kubrin and Nielson 2014 above by demonstrating the prejudicial effects of rap music in particular.
So, this bill addresses a real harm by forcing courts to seriously weigh the prejudicial effects. Courts can - and will - introduce creative expressions (including rap!) into trials. But, prosecutors will be incentivized to have harder evidence in hand instead of making cases via that route alone, and furthermore courts will have to more seriously think about the a) social context of the creative expression (e.g. how confident should we be that this is specifically relevant to the crime at hand?) and b) the prejudicial potential (which we can see is heightened for rap music).
This is why there are so many news stories about victims of violent crime who unwittingly wandered into dangerous first-person shooter video-game tournaments, and rappers are renowned worldwide for their upstanding citizenship and good moral character.
It's actually difficult to find a news story about an "aspiring rapper" that doesn't involve violent crime in some way...
Said people about Jazz in the 20s, Blues in the 40s, Rock in the 50s/60s, Metal in 80s and early 90s, and rap music 90s to now. Next thing you know, one Dua Lipa will be associated with a Triple Homicide and Bad Bunny will be the Worst Bunny.
More seriously, criminal trials are designed in the US to establish beyond a reasonable doubt that a crime was committed. That someone is a rapper (aspiring, amateur, or professional) should not be used in that process precisely because of your diatribe about moral character: it has nothing to do with the person at hand, and everything to do with your eager use of the ecological fallacy (https://en.wikipedia.org/wiki/Ecological_fallacy).
If a far higher percentage of rappers have violent pasts compared to non-rappers, it's only reasonable to start with a higher prior of them having committed another violent crime. Evidence is still needed, of course, but less evidence than if they were not rappers.
By precisely the same logic, you would be ok with a 9x higher prior of committing a violent act conditional on being male, such that "evidence is still needed, of course, but less evidence than if they were [female]." Or, you would be ok with an 5-9x higher prior (depending on research) of committing sexual violence similarly conditional on being male. After all, the ecological fallacy is 'only reasonable'.
There's a reason that there is a bar for evidence - not everything that is (or is not) probative is admitted as evidence in a criminal trial. If you'd like to reform the criminal justice system to remove that bar, go for it. I myself prefer a system that evaluates evidence for the crime itself, rather than deducing criminality by way of the defendants' characteristics (except when those include specifically relevant information, like prior conviction for a similar offense).
I'd be OK with that, yes. Why do you assume it implies lowering the bar for convicting men, rather than raising the bar for convicting women? In practice, the jury have a good intuition for what is or is not plausible male/female behavior based on their wisdom, accumulated through their collective lifetimes of experience. This intuition is reflected in how much evidence they need to convict. It's Bayes' law in action.
Ok, I think that's a genuine preference divide between us then. I prefer a higher bar than you do, because you seem more trusting of CJS dynamics than I am. Oh, and I don't say anything about the absolute bar for men vs women - I talk about a relative (9x) differential. You could use that to create a higher bar for women, or a lower bar for man, depending on your standpoint.
As for juries wisdom, color me still dubious. There's some evidence of bayesian reasoning in civil jury trials, but criminal jury trials have a pattern of problems:
Empirical assessment of actual jury trials:
Anwar, Shamena, Patrick Bayer, and Randi Hjalmarsson. 2012. “The Impact of Jury Race in Criminal Trials.” The Quarterly Journal of Economics 127(2):1017–55. doi: 10.1093/qje/qjs014.
Laboratory assessment of jury trial dynamics:
Lynch, Mona, and Craig Haney. 2009. “Capital Jury Deliberation: Effects on Death Sentencing, Comprehension, and Discrimination.” Law and Human Behavior 33(6):481–96.
Lynch, Mona, and Craig Haney. 2011. “Mapping the Racial Bias of the White Male Capital Juror: Jury Composition and the ‘Empathic Divide.’” Law & Society Review 45(1):69–102. doi: 10.1111/j.1540-5893.2011.00428.x.
Given the rap-specific research in my original post, and the juror process research here, I am confident that 1) a biasing effect of 'rap on trial' exists, and 2) that effect layers onto already-known jury-emergent biases. Hence directing judges to weigh the probative value of rap lyrics against those prejudicial effects.
Yea men should be treated very differently as criminal suspects than women (and are).
This in the context of a criminal trial, when presenting evidence to a jury (court decision making). NOT in an investigatory context (law enforcement agency decision-making).
Harsher treatment of male suspects, esp. for sexual violence is already what happens. Why would rappers need special carve out?
The big question in both cases is "to what extent is this circumstance independent of the police arresting the suspect?".
That is to say, if men commit 9x as many violent crimes as women and also get brought to court for violent crimes 9x as much as women, then (assuming similar proportional rates of getting away with it) the defendant being male is zero evidence of guilt; it overlaps rather than stacks with "this person has been arrested and charged with a crime".
I'm not sure how well "is a rapper" and "is male" are already factored in by police investigation; I imagine both are to some extent, but whether it's less than, the same as, or indeed more than the base rate (the latter case being a scenario where e.g. rappers commit murder 9x as often but are arrested for murder 18x as often, so the defendant being a rapper is actually evidence of innocence) I don't know.
I recall reading one of the big official studies on rape and there being some massive multiplier, but this being entirely due to the fact that they didn't count "woman forces man to penetrate her" as rape; once I manually included forced-to-penetrate as rape and recalculated the headline numbers it was very close to 1:1.
It is not, however, rational to assume that jurors in a criminal trial will be Bayesian rationalists. "Priors", are not really a thing in this context.
Sure, but “let’s not show some kind of evidence to juries, because jurors don’t know in general the right way of interpreting evidence” seems kind of iffy in a legal system predicated on (among other things), guilt being established via trial by jury.
I know there are already (lots of) restrictions on what kind of evidence you are allowed to present to a jury, and I even agree that a lot (maybe most) of those are good to have. I’m just saying, “jurors are not Bayesian rationalists” is a fully-general argument for doing *any* kind of evidence filtering. You should need more than that to just block whatever kind of evidence.
(To be excessively fair, the law in question does at least claim a bit more support than that, or at least different, but it’s not very convincing to me personally.)
Also, dionysus said “only reasonable”, not “Bayesian rationalists”. Those are completely different standards. (And, AFAIK, only the former is actually used commonly in judicial arguments.) The “reasonable man” might not use the word “prior”, but that doesn’t mean his reasoning process isn’t reasonably (sorry) described by it.
For example, when considering the case of an average adult human being beaten to death, a reasonable juror would find it more likely, before looking at the evidence, that it was done by a big, strong man might than a by tiny teenage girl, even if they won’t use the term “prior” to explain their reasoning.
I don't think so. All evidence should be related to the crime instead, not the fact that the alleged perp portrayed similar crimes. The world is full of great literature that is full of despicable criminals. That is neither here nor there. Anyone, in order to be convicted, needs to be convicted based on evidence of that individual having actually performed that actual crime.
However, I feel that is so obvious and clear within the context of present law that a bill like this is unnecessary.
None of this a is about someone BEING a rapper, it’s about the content of the rap.
Which is why the ecological fallacy has nothing to do with it.
Edit -- see above, I was responding to SlowlyReading's ecological fallacy, so what I say below is superfluous
Yeah, in theory this is correct -- specific content should be at issue, not the status of being a wrapper. That theory breaks down in application, however. For specific case examples (I see you are a retired lawyer), see Dennis 2007:
Dennis, Andrea. 2007. “Poetic (In)Justice? Rap Music Lyrics as Art, Life, and Criminal Evidence.” https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1104756
The reason I believe the ecological fallacy applies is that the pattern of inclusion of rap lyrics, and the evidence of prejudice, show that prosecutors are establishing culpability (and sometimes mens rea) by BOTH the lyrics and the identity of the defendant as a rapper. To consider this, think of the counterfactual: recent (90s or sooner) cases where defendants were convicted on the basis of lyrics that were NOT rap. One could easily imagine metal genres supplying such a counterfactually violent lyrics. In the absence of that counterfactual, the presence of many factual cases of rap lyrics being used, and the prejudicial effects of the 'rap' aspect of otherwise controlled lyrical expressions, I argue that there is an ecological fallacy being applied.
🙏
Oh, in the context of this comment-reply chain, I was responding to user SlowlyReading's comment above:
"This is why there are so many news stories about victims of violent crime who unwittingly wandered into dangerous first-person shooter video-game tournaments, and rappers are renowned worldwide for their upstanding citizenship and good moral character.
It's actually difficult to find a news story about an "aspiring rapper" that doesn't involve violent crime in some way..."
Invoking a class of people's (rappers) 'lack of renown for their bad citizenship and poor moral character' as a reason to punish INDIVIDUAL members of the group for SPECIFIC alleged crimes is a classic example of the ecological fallacy.
Ah, yes, news stories. Gell-Mann amnesia.
Find me a single news story about a chamber music concert or video-game tournament that ends in violence, the way that news stories about violent aspiring rappers appear in every metro newspaper every single weekend.
https://pittnews.com/article/134156/state-national/several-slain-in-mass-shooting-at-florida-video-game-tournament
I mean, since you asked. Yes, it's just one.
Roasted.
List of classical music riots, though most sound pretty tame:
https://www.thechatner.com/p/classical-music-riots-ranked
Well, as a baseline, news stories about people in general tend to involve crime...
As Tim McCormack says, news stories are very often about crime. And if you're arguing that it's not unfair to use rap lyrics as evidence of criminality, what matters is not whether news stories about "aspiring rappers" _involve violent crime in some way_ but whether they involve violent crime _in the specific way where the aspiring rapper commits the crime_.
So I just did a search (with DuckDuckGo, if anyone else wants to check) for "aspiring rapper" on the CNN website. After eliminating duplicates and stories that didn't actually involve an aspiring rapper, the first 10 results I found had _two_ cases where the aspiring rapper was a violent criminal or somewhat-credibly alleged to be one. (The others: 3 murder victims, two sexual abuse victims, one person criticized for using the word "n----r", one QAnon event organizer, and -- kinda ironically given the context here -- one student suspended from school for writing a rap accusing the school's atheletic coaches of sexual harassment.)
Instead of CNN, why not search Google News or some other aggregator? National media is much more determined to conform to The Narrative than are the local news outlets, whose reporters actually have to share a community with aspiring rappers and the misery they inflict upon it.
Because CNN was the first US news website that came into my head. (Why US? Because rap is primarily a black-American thing, and because most ACX readers are in the US, so that seemed like the right space to be sampling from. Why the first site that came into my head? Because that seems like the simplest way to avoid either the appearance or the reality of cherry-picking.)
I had a quick look at Google News. Filtering out duplicates and things that aren't actually stories about specific aspiring rappers, I get: Accused murderer 2. Car thief 1. Drug dealer 1. Fraudster 1. Murder victim 2. Possession of a machine gun 1. Just a rapper (story about their music) 2. So, once again, only 20% of them have the rappers as perpetrators of violent crimes. The proportion of perps to victims among crime stories is higher here than for CNN; I suspect this is just random variation, but it could instead be as you suggest that different news providers have different biases and preferred narratives.
Anyway. Even with the absurdity of counting being a _victim_ of violent crime the same as being a _perpetrator_ of violent crime, I remark that both of my 10-story samples found several stories in which no violent crime was involved, and at least two where no crime was involved at all. So obviously your original statement that "it's actually difficult to find a news story about an "aspiring rapper" that doesn't involve violent crime in some way" was not only false but _trivially seen to be false_; I suspect it was "bullshit" in Frankfurt's technical sense of a factual-looking statement made without the slightest concern for whether it's actually true or false.
So I don't think you're really very well placed to complain about minor suboptimalities in my choice of how to check said claim. At least I made some attempt to do so. Nor do I think you're in a good place to complain about journalists' concern for "The Narrative" since it seems plain that you're engaged in narrative-spinning rather than truth-seeking yourself, on this occasion.
I almost completely agree with you, but SlowlyReading’s exact words were “involve violent crime in some way.” It never says that the rapper in question is the perpetrator.
I agree that his implied argument points in that direction, but still, if we go to the effort of verifying a claim we should actually verify the actual claim that was made, and do it right.
You did actually verify the actual claim (and just out of curiosity, I tried it myself, and the third “aspiring rapper” story I found on Google News did not involve any violents act that I could find). But you did expressed your disagreement mostly towards a different argument than the one you were arguing to.
OK, so reading back on my answer, the summary is “you did very well, but you could have done great”. Feel free to ignore me, it sounds nitpicky even to myself.
My only nitpick is that above I already acknowledged that SlowlyReading said "in some way" but pointed out that what he actually needs for the argument he was making when he said that was in fact more than "in some way", which is why in my comment immediately above I said "even with the absurdity of ...". So yes, I focused mostly on a different claim from the one he literally made, because the one he literally made is obviously irrelevant to the actual point at issue.
Also, the example is about the SEC, and a California evidence statute wouldn't apply in federal court.
Good analysis. As I suspected, the law seems to be a mild corrective to a probative/prejudical balance that got a bit out of whack. I'm not sure how many innocent people are in jail who would not be but for their violent lyrics, but cleaning up glitches in the rules of evidence seems like a good exercise.
It's a phyrric victory for gangsta rappers, though. I can think of no genre where authenticity is more prized, where there is less of a distinction between speaker and artist. Johnny Cash could sing about shooting a man just to watch him die, or doing cocaine and killing his old lady, or merely spending some time in prison, and no one judged him to be a fraud simply because he never killed anyone or spent any significant time behind bars. In rap, your boasts may be exaggerations, but if your lyrics say that you grew up poor, slung rock, and nearly killed a fool for... I don't know, frontin' or steppin', then those had better be more true than false. (The specific tale of getting stuck up at a dice game and rescued by one's homey Nate didn't happen word-for-word as Warren G described it, but it's a plausible embellishment of things that did or could have happened to Mr. G and his cohorts).
CA 2799 might as well be called the Okay Seriously Though We're Full of Shit Act of 2022. It _should_ cause a dropoff in boasts about felonious tendencies because some of the big names in the genre have admitted to being frauds. We should no longer presume that, after MC Whatever announces that he is "here to say," he actually does this or that "in a major way." We should presume that he does it in at most a minor way, if at all. But it will likely increase the boasts a bit, as they come with fewer potential criminal consequences.
I like your response but can't help but provide you your wish for a genre that values authenticity more: Narcocorrido! See https://en.wikipedia.org/wiki/Narcocorrido
I don't think authenticity (as far as VIOLENCE or THREATS goes) matters as much as the "grew up poor, slung rock" aspects, but I'm not an expert in rap authenticity and semiotics. As far as rappers whose threats and shoutouts are 'at most in a minor way', when I was younger I was a huge fan of Canibus (rapper name, not a misspelling). How can you not love the skill of someone delivering this live at a radio station? https://www.youtube.com/watch?v=LN_yJSMYdBg&t=424s
"Here's an optimistic mindset, if you late for today, just say you're early for tomorrow..."
I've never understood if by brain carves out a special exception for Canibus in my general "Dislike Rap" heuristic due to his skill (the rhymes of yesteryear just seem so much more clever than today, I'm sorry), or due to being linked to beloved comedy __Office Space__. Damn, it feels good to be a gangsta...
I remember there being some popular rapper with a quasi-criminal persona who turned out to have been a prison guard. People did make fun of him for that, but I don't think it ended his rap career.
Rick Ross
The Swedish approach to rules of evidence appears workable for them, it's hardly obvious to me that our system is better:
https://www.overcomingbias.com/2019/10/rules-of-public-evidence.html
Nah, it's just a way to ensure rapper tax $$$ -- a great deal of money, these days -- stays within California's juridiction by representing to aspiring artists that the culture of violence within that community will remain a bit more tolerated than juries, acting on their own judgment if allowed (yikes!) to see all the evidence, might otherwise permit.
To understand the "why" of how laws get enacted, follow the money, you won't often be disappointed. The idea that the California Legislature might enact law on the basis of some super-clever-subtle measurement of social science research[1] instead of what maximizes tax revenue and votes is a good joke, though. Well done! Someone who is unfamiliar with California politics might actually buy it.
-------------------
[1] Itself a paragon of reproducible rigor, right?
If you'd like to claim that the research I've cited is specifically false or unreproducible, I invite you to clearly do so and explain why.
Otherwise, thank you - you've been invaluable for helping complete my AstralCodexTen comment section bingo card (if nothing else)!
Nah, sorry, that's not the way it works in science, or for that matter in real life. It's the person raising the extraordinary claim -- in this case, the ridiculous and offensively snobbish claim that ordinary adult men and women who sit on juries would be utterly bamboozled by art, unable to distinguish between the role an entertainer plays on stage and what he's like in real life -- which requires strong proof. And a random paper or two in a social "science" journal is a laughably inadequate attempt at that.
Next up, you should persuade the California Legislature to prohibit the jury in the Ron Jeremy case hearing that he was a pr0n star, because OMG they couldn't possible distinguish between what he did on camera and what he did in real life. Right? And then there's Alec Baldwin -- better not let any jury hear about the fact that he played a gunslinger on film, if he's to get a fair trial on the Halyna Hutchins case, if he's charged. How can you expect mere mortals to distinguish between Baldwin's character on film and his character in real life?
Oh wait, what's that? You say the Valley pr0n industry is no longer what it once was, doesn't provide a nice stream of tax revenue to the California Treasury, and so your interest in zealously guarding the rights of pr0n stars charged with crimes is....lower? So surprised.
🤡
Trivial warning (1% of ban): Low-content, high-temperature comment.
To Scott Alexander: The rules as I read them are to be considerate and truthful, or at least one. I have been considerate and truthful throughout, except when faced with the most clownish of comments, and then fallen back on truthful.
Context: Numerous productive comments by me, and then Carl claims that it's unlikely that the Californian legislature would make a decision on the basis of science. This is despite my top-level comments quoting the legislation, which explicitly references such science. Instead, Carl Pham proposes the 'argument' that: "Nah, it's just a way to ensure rapper tax $$$ -- a great deal of money, these days -- stays within California's juridiction [sic] by representing to aspiring artists that the culture of violence within that community will remain a bit more tolerated than juries, acting on their own judgment if allowed (yikes!) to see all the evidence, might otherwise permit."
I respond that "If you'd like to claim that the research I've cited is specifically false or unreproducible, I invite you to clearly do so and explain why. Otherwise, thank you - you've been invaluable for helping complete my AstralCodexTen comment section bingo card (if nothing else)!" Otherwise, it would just be repeating the cycle: broad claims, with no basis, that everything I say with evidence is wrong. There's no point in such a discussion, because I'm playing according to rules that the other party will not.
Carl Pham's response confirms my suspicions: the first paragraph just doubles down on the same tired canards:
"Nah, sorry, that's not the way it works in science, or for that matter in real life. It's the person raising the extraordinary claim -- in this case, the ridiculous and offensively snobbish claim that ordinary adult men and women who sit on juries would be utterly bamboozled by art, unable to distinguish between the role an entertainer plays on stage and what he's like in real life -- which requires strong proof. And a random paper or two in a social "science" journal is a laughably inadequate attempt at that."
This is clownish for several reasons, noted also by smilerz's incredulous response to Carl Pham: Science actually does work by people criticizing specific things (and then hopefully ameliorating them). As to the studies? Carl Pham will not actually criticize them substantively, and instead just says they are, in totality, "laughably inadequate".
Following this, Carl Pham proceeds with two paragraphs of disturbing/weird non-sequiturs about 1) ron jeremy, 2) alec baldwin, 3) 'the Valley pr0n industry', and 4) California treasury. I would argue that clownish... is an understatement.
I stand by responding with a clownface emoji to a hopelessly clownish comment. Otherwise, if it looks like a duck, quacks like a duck, and tastes like a duck, I'll be forced to claim it's foie gras. However, I recognize that emojis are nasty, short, and brutish tools: more apt to group chats than respectable substacks. Therefore, I am open to a safeword that substitutes for a clown emoji as a way to identify hopelessly clownish comments; perhaps 'non, est tibi'?
To Carl Pham, I am sorry about using a clown emoji to characterize your comments. Now, and in the future, I will respond to clownish comments by you with "non, est tibi".
The extraordinary claim substantiated with multiple studies?
The way that it works now is that you have to criticize one or more of those studies in some specific, meaningful way.
The idea that the California legislature enacts law on the basis of maximizing revenue seems highly dubious to me. My guess is that deadweight loss abounds and if they asked some economists how to improve things there would be plenty of suggestions, which they'd have little interest in enacting.
Yeah, California isn't known for good policy. Prop 13 is, well, a ballot proposition and therefore not the legislature's doing, but should set a fairly low bar for the rationality of the state.
It's a highly dubious proposition to you that people might act in such a way as to maximize their self interest? Or it hasn't occured to you that the options of Legislators for doing interesting things that bring them fame and power rise and fall with the amount of tax revenue available with which to fund them? Either is a fascinating lacuna in common sense.
I don't model state governments like a profit maximizing business. They are far more irrational & dysfunctional. They will enact laws & regulations with lots of deadweight loss for no good reason. An economist suggesting they replace rules with fines/fees that could at least generate revenue would get ignored.
What brings them fame and power is whether the laws get good press, not whether they actually improve the economy by 0.01%.
This all makes sense if we're talking about the general use of violent rap lyrics as a reflection on a suspect's character and likely criminality. But there have been mutliple cases involving *specific lyrics* seeming to align with the *specific criminal accusation*.
I have an amateur interest in laws and regulations (they are like programming in natural language), specifically their wording and logical structure, and how it affects their interpretations.
I have a question about
> a form of creative expression
What *is* that? Is a cringe tiktok sketch a form of creative expression? Is a reddit AmITheAsshole threads? Does there have to be an audience? How much? Would it matter if the audience is anonymous? Does it have to be recorded? What if I admit to police officers I'm guilty but I later claim that I was just trolling for the views? (and would it matter if my phone recorded the admission or not? Would it matter if I posted it to my 10K subscribers youtube channel and people laughed and took it like a joke?)
It sounds very naive to me to weaken a piece of evidence based on, essentially, the answer to the question "is it art?", the question is literally a meme because of how useless and impossible to resolve it is. I hope that I'm the naive one and that CA courts actually have a concrete list of things that are "forms of creative expression".
I actually have no problem with rap, a small minority of it is fire, the vast vast majority is utter garbage sang by wannabes who never graduated from the "I foked yo mama" mentality, but hey, Free Speech rules, every dumbass has a right to say every single dumb thing that comes on their dumb mind, and the world would be a pretty dull place if every single word I say publicly can be used as evidence against me arbitrarily far into the future.
The comparison to other styles of Art is weak to me, GTA is pretty obviously about a bunch of made up triangles on the screen, no matter how much you give them human names and human faces and depict violent crimes and other bad things happening to them, it's never going to match a rapper talking about fucking his opponent's woman. Because the rapper is a real guy talking about another real guy's real woman. This can get pretty murky if/when we get VR/AR/Brain-To-Brain fiction, but right now the difference is very clearly cut, GTA's characters are entirely a fiction residing in computer memory, it's very safe to do whatever you want to them.
I find the bill absurd, mainly because it's so unnecessary. Who in the world imagines that an artist of any kind depicting violent crime must be a perpetrator or even an encourager of violent crime? Yes, some rappers are violent, as are some writers of acclaimed fiction. But most are not. Why isn't this obvious?
At a book discussion I was in lately, someone said she hated the book because she hated the main character. I have read books that I love in which every character is despicable (they abound in work of Jim Thompson); so what? Thompson himself was not a despicable character.
Some years ago, a black kid was arrested in NY for forcing a subway token clerk to hand over money by placing a lit rag soaked in some solvent under the window. There had recently been a movie where such a crime was described. However, it turned out that the kid had never seen or even heard of the movie. He was duly convicted, but on the evidence at hand, not on the basis of his having seen the film. (Actually, I'm not sure the belief that he must have seen the film was raised in court.)
Though my examples are not about the exact situation that the bill addresses (the culpability of the artist him or herself), the same principle holds. Evidence has to be about the specific crime, not about the situations the artist has portrayed.
And if you're going to pass a law like this, why should it be limited to rappers? Should apply to any artist or writer accused of being guilty of a crime the artist portrayed.
He provided multiple studies showing evidence of prejudicial effects, in what way are these studies flawed?
You also have to pair "unnecessary" with "what's the possible harm?" I have a hard time imagining any substantial harm in forcing government to work harder in connecting artistic content to actual crime.
The further wrinkle for kidney donation is that when somebody altruistically donates a kidney, they have demonstrated reduced left kidney function, so they would probably insist you donate the right one instead.
What is this about Thiel and democracy? Did he lose an election for kindergarten class president or something?
Oh boy, mindless pathologization of our political opponents, how fun. I guess democracy cultists were people who were losers in kindergarten and individuals having more social authority than them hurts their feelings.
Democracy us when billionaires donate to my team, and when they donate to the other team it's "undermining democracy"
Not sure if I should feel happy or sad that the house parties I attend mostly revolve around drinking beer and determining what distance from the fire will bring the temperature to a tolerable level
Cool, another reason to work on upper body strength so I can hold on to the rope and not fall to my death.
Also, that rap is the hottest yet!
> So if you donate your evil kidney, then you become entirely good, but the recipient becomes entirely evil.
Except that when someone qualifies to receive a kidney, the doctors transplant the donor kidney into the recipient, but they LEAVE the old kidneys there too. So actually the donor becomes 100% good, and the recipient only becomes 67% evil. This changes the whole moral calculus.
Excellent point
What if you put the new kidney on the right side? Does it become good? Or just evil and confused about its location? So many questions.
There are people whose organs are mirrored in their insides. Do they also have reversed kidney goodness/evilness?
But do failed kidneys still maintain their good/evil capacities? Or, having ceased to properly function, do they become morally neutral?
Yeah but does the potency of the good or evil essence decrease with the functionality of the kidney?
"I just heard about STARS, I think it's a hot stock tip"
"Whoa! Hold up! You can't just go saying it!"
[record scratch]
"If they're not publicly traded, you have to be an accredited investor,
A million in wealth, or large income for a couple semesters
Are there conflicts of interest? Then that's right out.
Talk to your trusted lawyer if there's any doubt.
Now we have to figure out the finance, don't issue convertible debt
Most founders prefer SAFEs, that's something you shouldn't forget!"
[mic drop]
You sigh and walk out of the room.
This should be performed at the next YC induction.
My best dumb start up idea is a toilet paper dispenser that rolls out directly from the cash register as you scan your purchases proportionate to how much you will probably need for that item. The roll drops into your bag automatically as you leave. No name for it yet but Make-a-Poop-take-a-Poop seems like it would never fly.
So funny! You know, I read a "bay area house party bingo" board meme recently and thought, "those are all Slate Star Codex topics."
> See, that’s the kind of negative talk that used to get me down. I would have given up. But now I just think to myself - did Jesus give up when the Minotaur kidnapped his daughter? No! He set out through the Fiery Forest to find the magic helmet that would bring her back! And that’s why I’m not going to give up either. See! I’ll let you have that one for free.
Quick, somebody DALL-E this up!
Eh. I'm not too proficient with dall-e, but it tends to write too short of a text to really develop things. Still...
Prompt: Write a fairytale about Jesus trying to find his daugther that has been kidnapped by the Minotaur:
Result:
The Minotaur was a fearsome creature that lived in a labyrinth deep in the forest. Jesus's daughter had been playing in the forest and gotten lost, and the Minotaur had taken her. Jesus searched for her for days, but he couldn't find her. Finally, he came to the labyrinth and confronted the Minotaur. The Minotaur was about to kill Jesus's daughter, but Jesus fought back and defeated it. He saved his daughter and they went home safely.
Another attempt:
Once upon a time, Jesus' daughter was kidnapped by the Minotaur. Jesus set out to find her, but the Minotaur was always one step ahead. Jesus searched high and low, but he couldn't find her.
One day, Jesus had a dream. In the dream, he saw his daughter. She was in a dark cave, and the Minotaur was standing over her. Jesus woke up with a start. He knew where his daughter was.
Jesus went to the cave and confronted the Minotaur. He fought bravely, but the Minotaur was too strong. Just when it looked like the Minotaur was going to win, Jesus' daughter ran into the cave. She distracted the Minotaur long enough for Jesus to defeat it.
Jesus and his daughter were finally reunited. They were both happy and safe.
The output of Dall-E 2 is images. The output of GPT-3 is text. This looks like text, not an image. Watson concludes that this was more likely to be generated by GPT-3 than by Dall-E 2.
DALLE-E is an image generator
Loved the Wikipedia section! Whenever xkcd mentions Wikipedia, I immediately go and check the relevant page for joke edits. This blog probably doesn't have the same reach and hopefully isn't read by Wikipedia jokesters, but I'm going to be watching the Hofstadter page for a while, just in case.
And just a few sentences later in that Gemara it says " And Rava said: The world was created only for the sake of the full-fledged wicked or the full-fledged righteous; others do not live complete lives in either world. Rava said: One should know of himself whether or not he is completely righteous"(Berachot 61b), so by donating his kidney, the donor know his status, and knows that the " completely righteous are written and sealed immediately for life; the completely wicked are written and sealed immediately for death; the beinonim (average) are held in abeyance." (Rosh HaShana 16a). Seems like a good shortcut to getting into the World to Come.
Hahahaha
I predict that the next time Michael Levine goes to a house party everyone’s going to want to talk to him about rapping to avoid getting convicted of security fraud.
I'm going to assume you're making a joke about his statistically likely name, but for those who are confused: Maxander is referring to Matt Levine, author of Money Stuff.
But what’s the solution for save-state ear worms? Ear worms that sing the next part of a song that has been paused right before resuming the song.
I haven't worked my way back to Bullfinch yet. I'm still stuck on "The Golden Bough". Sacrifice of our sacred king forestalled.
I haven't had a social interaction in 23 days and this only makes me happier with that decision
And to think there is only a 0.01% chance Russia nukes San Fran this year.
"Did Jesus give up when the Minotaur kidnapped his daughter? No!"
That's brilliant. Were you thinking of Animal House?
Bluto: "Was it over when the Germans bombed Pearl Harbor? Hell no!"
Otter: [to Boon] "Germans?"
Boon: "Forget it, he's rolling."
It's worth noting that CA 2799 doesn't actually ban the use of rap lyrics as evidence. Ken White (Popehat) has a Twitter thread on it here: https://twitter.com/Popehat/status/1575997780853551104
To sum up, the law basically just says to judges, "Hey, before admitting art by someone as evidence against them, consider the genre conventions, consider that it may not be describing crimes that actually occurred, and that if there's no reason to think that any crimes described could relate to this particular case, then the evidence might be more prejudicial than relevant." So, it doesn't actually change very much, as judges already consider questions of prejudice vs relevance all the time; this law just gives them an explicit instruction about some factors to consider. But, y'know, they're judges -- if they don't like the instruction, what are you going to do?
Q: Why did James and I leave Google to start a gambling startup?
A: There's no alpha left in Alphabet
Without the alpha, you did not want to get stuck with the bête?
There's no beta left in video tapes.
There's no pi left in pumpkin.
There's no mu left in dead cats.
There's no delta left in the limit where epsilon approaches zero.
Edit: There is no tau left in the set of taus that can be named
This would be funnier if the meaning of alpha wasn't quite well-defined, from finance.
> "I definitely don’t think you can make life have extra meaning just by making more myths"
Says you, Uriel.
Uriel's the one who literally added a book to the bible to patch the universe, he would TOTALLY be on board with doing that.
CHANGELOG:
Myths are still causing meaning creation; deprecating the mythology subsystem, to be removed once the legend subsystem is stabilized. New users may also consider the experimental third-party “Hollywood” module.
I was inspired by your previous house party post and started hosting house parties in Singapore. So far it's been great, the only problem is a skewed gender ratio (not enough men).
I'm unironically the AI skeptic guy. Think about it: we already have machines that can make cappuccinos at least as well as your average human for maybe, what, a dollar? But we still pay humans 4-6x times that to make one for us anyway, and sometimes we feel so guilty about underpaying that we actually TIP!
The difference is that the cappuccino guy makes it right in front of you (you're paying for the show of it) whereas an intellectual worker normally just delivers the finished product of their mind, and therefore can be replaced by computers without anyone being able to tell the difference.
Exactly. We attach value to inconveniencing others for our own benefit. It's fundamentally an expression of status and hierarchy, which capitalism conveniently commodified and made available for sale to anyone with enough money. Inconveniencing a computer just doesn't do it, we need humans.
It's the "low touch" business model, which works fine. See RyanAir, Aldi, and coffee vending machines.
I unironically want the address. Fork it over.
EDIT: Uh, please.
"Exactly. We attach value to inconveniencing others for our own benefit. It's fundamentally an expression of status and hierarchy, which capitalism conveniently commodified and made available for sale to anyone with enough money. Inconveniencing a computer just doesn't do it, we need humans."
But that is exactly what I was trying to say. I guess I didn't say it clearly.
We pay for the show of coffee being made, which proves that someone is working for us. As you said, it is an expression of status.
But none of that applies to intellectual and artistic work.
It used to be that wealthy patrons could commissions artwork done, which proved their status. That is no longer the case in our age of easy reproducibility of pictures and sound, but at least you know that someone worked on a piece of art. Right now, I can look at a picture or read a novel and imagine all the human work that went into it, even if it is not done specifically for me. Soon even that will no longer be true.
Wealthy people commission artwork all the time. Human effort still has its own intrinsic value. The automated stuff is handed off to the proles for their low-value consumption.
So will it be for AI. That shite self-published novel that is sold on Amazon as a print-on-demand paperback will get an AI cover made by stable diffusion and ripped off for free, but no one with any money or status will settle for something like that.
Right now, we "proles" are still getting the fruits of human artistic efforts, not the automated stuff (obviously, since there isn't much truly automated art yet).
If in a not distant future the masses will no longer be getting human made art (because there will be no way to tell it apart from the AI made stuff which will be if anything superior), that sounds like an immense and unsettling change.
Perhaps a little man-made art for a few extravagant elites will survive (and I doubt even that, since it will be hard to prove that something is man-made), but that is cold comfort.
I'm not sure how much you can blame capitalism for this one; modern restaurants really came into being in the aftermath of the French Revolution when the former kitchen & tables staffs whose previous employers were abruptly several inches shorter started providing their services to the general public.
Blame? I'm not saying it's a bad thing. If anything, capitalism democratized status; it allows everyone to at least have a turn enjoying a sense of status relative to others. Serfdom is a life of uninterrupted subjugation, whereas at least a wage slave can feel like a big man wearing a piece of branded clothing or driving a lifted truck. People roll their eyes at that sort of thing but status is a fundamental human need and while hierarchical status is a zero sum game, status symbols are basically infinite.
That's not true universally (drive thru Starbucks being one example) and, arguably, the reason they DO make it right in front of you is to PROVE that it's made by a human.
Anyway, it's pretty clear AI is just a new art form, and the artists/artisans (programmers) will capture the value created by their AIs, just as photographers captured value previously captured by painters, and the painters went off to do something else.
That something else turned out to be postmodernism, which creates ugly nonsense that a five year old would be ashamed to call his own and sells it for millions of dollars, all dressed up in leftist woo-woo.
The introduction of photography predates postmodernism by like a hundred years.
But after 1914, what painters went off to do was of significantly lesser aesthetic value as before and now we are significantly worse off from the artistic point of view since photography is not in the same league as the great painters of the past. Effectively we have lost a form of art
I promise you, oil portraiture is still a vibrant art form with plenty of money and talent behind it. The reactionary narrative of culture degeneracy definently gets people good and agitated but it's totally made up. It's fun to speculate about how much of that narrative literally traces back to Hitler getting rejected from art school.
People who pay $4-6 for a coffee that's no better than a machine-poured one aren't doing it because they enjoy having someone make it for them, they're doing it because they can't tell the difference between good and bad coffee and are cargo-culting some of the accoutrements of what they think good coffee supposedly looks like.
This is the Starbucks business model: sell bad coffee, in a way that makes it _look_ like good coffee, to people who can't tell the difference.
(Personally I _am_ one of the people who can tell good coffee from bad, but I'm not going to be too precious about this because I know there's other things I don't care about; for instance, I don't really care about wine. I could probably be fooled into drinking (and paying for) the Starbucks equivalent of wine.)
The quality isn't entirely in the roasting and the beans, that's only part of it.
The hardest part to master is getting your foam right. Starbucks sure as heck can't do it, which is why a Starbucks cappuccino is a wad of thick foam sitting on top of an espresso shot. You tip the cup, the espresso goes down your throat, and the foam stays where it is. It's a skill to use a steaming wand to get foam that's the right temperature, texture and consistency all the way through, then to pour it on top of the coffee in a way that mixes coffee and milk appropriately (latte art isn't just for showing off, it demonstrates that the two components are mixed the whole way up the cup).
Even just pulling an espresso shot is an art, because you have to know when it's had enough. If you under-extract or over-extract then you get a bad-tasting coffee. Good baristas know when to stop by watching the colour of the mixture as it comes through. (Starbucks baristas press a button which extracts for a consistent fixed amount of time, randomly over- or under-extracting the beans depending on other factors.)
I'm guessing you have no experience making espresso before?
Then you're not really qualified to be confused about the existence of baristas.
>Think about it: we already have machines that can make cappuccinos at least as well as your average human for maybe, what, a dollar
A barista's coffee making ability is significantly above that of the average human. The average barista made espresso is noticably better than the average consumer pod machine.
I knew where the Hofstadter story was going... and it still made me laugh a lot
I always really enjoy these Authentic Bay Area Experience™ writings.
This makes me glad I don't have a social life 😁
Hmm and yet isn't 'this' (commenting on ACX) part of your social life?
Only if he answers
At some level I'm just completely unable to tell how much is satire and how much is reality.
This is the reality part: “A few people look up, mildly alarmed, then decide that it’s probably just some new kind of trad heterodox reactionary thing they don’t want to know about.”
As an old maths teacher of mine once said to me, "you and I both know that between doing maths and having a social life there is a choice to make, and that we both made the right choice.
This was great, thank you for sharing it! Always a delight to read your short stories.
> “Resisting the urge to go to events like this”, you avoid saying. “What about you?”
So what *did* the protagonist answer? Or was there just an awkward silence before the protagonist passed the ball back to the asker? :D
> All eyes turn to you.
> “Wait,” you say. “I thought urbanism was about finding ways to make cities more liveable and affordable! Not about trying to ban all the things that make residential life convenient and pleasant!”
I feel a little like these two paragraphs got swapped... or I just missed what triggered all urbanists to look at the protagonist.
"Resisting the urge to stay home alone", probably.
Bruh is this art?
when are you uploading your rap to soundcloud?
"Procedural myth generation" is old hat. See every pulp hack who got into the game after Lovecraft/Howard. Generate random letters for names of incomprehensibly evil deities. Roll dice for numbers of tentacles and eyes. Add muscle and sex allusion to suit. Mythos complete, put in the mail, rinse, repeat.
Then of course the '80s Tolkien clones full of small people racially classified as something ending in "it" or "ing," allied with tall, slender, pointy-eared people called [random letter]['-elf" or "elve"]s" against the Ultimate Evil Enemy (break out random letter generator again).
And yet, the popularity of these suggests they're hitting *something* in our souls/evolutionary programming/etc.
Oh, I didn't mean to suggest otherwise. In fact, I'd say that the laziness/ease of tapping into that with minimal creativity demonstrates how deep-seated the archetypes or whatever are.
Nuglush was a forgotten beast. An enormous eyeless lobster. It has large mandibles and it undulates rhythmically. Its white exoskeleton is warty. Beware its poisonous sting!
Nice, I was wondering if anyone was going to bring up Dwarf Fortress' myth generation.
In some other closely-aligned universe branch, my bookshelf runneth over with delightful tomes from acclaimed novelist Alexander Scott, after his blog !Song failed to find an official publisher.
Stupid startup ideas: well, more like a specific product rather than a startup (but I guess some startups do revolve around just one Novel New Product). But many years ago, I got frustrated with how clothes hampers are LIFO-designed. That's improper rotation of laundry! One ought to wash the oldest clothes first. So I ended up pitching the idea for the double-sided FlipHamper(tm) to one of those skeezy We Buy Your Invention-type companies. (For whatever it's worth, they thought it had promise, and could begin prototyping if I'd just send them three easy payments of $19.99...)
These days I don't generate enough laundry to make such a thing worthwhile, but I think the general concept could still be useful. Lotsa times in life where I want to access stuff on the bottom of a container, and it'd be super easy/intuitive to just flip it over. But, alas, there's only one opening. (Incidentally, this is the most efficient way to unload a box of bananas, because they're shipped upside-down. Easier to grab the stem ends.)
You mean the alternate branch where Alexander Scott got annoyed at some random blogger and wrote a nasty article about him in the NYT, and the guy had to move to the new Mars colony to get away from all the hate mail?
I think you missed a dementia joke. Something like "There's no alpha left in α-synuclein". (or maybe β-Amyloid)
+1
Remembering why I moved to Utah..... dyyyyying 🤣
Though now I feel inspired to write “another utah house party....” it’s a whole other kind of tech bro here 🤣
I would be interested in hearing more about how Utah tech bros are different to California tech bros, and it doesn't need to be in the form of a witty short story.
My first guess is that they all like skiing and talking about how glad they are that they don't live in California.