#genericize everything

[cutthroatpixie]

kind of obsessed with this ngl

no idea where i was going with this but i abandoned it at the most disconcerting moment possible

[amplexadversary]

I'm starting to get annoyed enough at all the *aesthetic* posts that post two of the same picture twice (to get "four" for, what, engagement reasons or some pseudostatistic?) that I'm very close to just reposting them correctly instead of reblogging them when I find one I'd otherwise like.

I'm going to tag this "fixed that for you"

Seriously though, you really couldn't get two more pictures? Post one on top of the other? Do anything other than something blatantly stupid-looking and lazy click-courting?

[felltyscrap]

I deeply hate the proportion systems of measurement i was taught bc they feel genericizing of people (ideal man 8 heads tall reeks of cockroach wife in all its bleakness) but all my shit tends to skew wildly as a struggle to "see" how something looks in its greater context and accurately make every line respond to it. I think some of this is related to fatigue but my hunch is i need to workshop a system of measurement that works better for me.

I feel like in general certain shortcuts feel quietly like I am ignoring the deeper life stiring in everything but my brain cannot keep up at all with how that factors onto a page. You can tell where I'm excited and where I'm bored/overwhelmed really easily, its a little like a kids drawing. Its been a battle to circumvent that tendancy its so strong in me hahaha. Maybe I dont understand that tendancy that well if i keep fighting against it.

It's just like...sort of shutting down and wanting to be done the moment the drawing goes out of wack. Theres a sensation i was "working on something" that gets awashed, a bit like playing the piano and messing up a note and then wanting to repeat it over and over again....but for me its like....i want to try it from a different angle instead of redrawing the things i was fine with. I nailed all these chord progressions except for this one so lets try the one i messed up in a completely different song- to keep it interesting.

I think uh. I like drawing because i enjoy the novelty of kinda sucking at something and grinding away until i get it and i can blast through things. I love....mindless movement based grinding I spent 800 hours in tf2 sucking badly at rocket jumping and brute forcing my way through beginner maps. I did not get much better and I am embarassingly bad at it but the thrill of landing a jump after about 30 minutes of failing over and over again is sooooo juicy. I loved celeste for similar reasons. It's just harder to give art that mindless grind bc i have a soft spot for a style thats really recollection focused. My memory game. My memory smoothie.

I think this style of learning tends to yield to extremely slow progress. I play with reoccuring things over and over but drop and pick them up really randomly based on whatever i can squeeze joy out of in that moment. Its less about learning and fundamentally about whats agreeing with me in the moment. It makes sense to me why the idea of illustrating for others or drawing a comic beyond a train of thought doesnt work- some days im into arms and other days im into jeans. Its highly contingent on what im interested in and if I dont follow that i feel like I'm being put in an awful box and I will rattle the cage and yell even if i tried to put myself there for the sake of growth.

This becomes worse when i'm unsure of whats interesting to me at the moment. I -am- really out of controll! But I have no idea how to ride this chaos besides drive, which i have in insane spades. I will rush out into the sea over and over again until i can ride the surf.

[velvetvexations]

(anyone who may know better - I happily welcome corrections!)

I was reading some more into the Second Sino-Japanese War, as is my wont, and I think the key to understanding Imperial Japan is that they were (A) obsessed with securing the dominance and propagation of Yamato culture via (B) copying everything the West did as closely as possible. They were undoubtedly motivated by a very strong envy of Western empires and entitlement to a seat at the same table. Not that they were more entitled than the ones already sitting at the table, of course! No colonial asperations were morally pure or justified beyond simple nationalistic greed. But it's interesting because in those other cases the table, by which I mean not just the benefits of empire but the balance of power where the Western empires had a stranglehold on the world, kinna gradually formed between them over the course of centuries, and Japan was like, on the outside trying very hard to break in and take that for themselves after the Perry Expedition. And holy hell did they go a long ways towards making themselves a modern world power in an extremely short amount of time, the exact opposite of the Qing Dynasty, which was in such disarray during this entire period that by the time WW2 rolled around it was interrupting China's ongoing civil war that had been going on since a couple of decades earlier when the Qing had fallen.

Like, they suppressed the local languages of their colonial conquests but then, also, look at this:

That's a group of IJ officials pretending to inspect where a fake bomb was set off to false flag a justification for taking Manchuria. Japanese people dressed in "Western"-style clothing is hardly anything surprising now, but in-context, it's strange. Because like, IJ was trying so hard to make all of East Asia distinctly Japanese, right? But it was all filtered through that desire to be as powerful and glorious as they saw the Western world. IJ was, in a sense, stuck in a mindset of "anything you can do I can do better" - towards the West, with East Asia just being the tool with which they proved that with.

Fear of a Qing-style Century of Humiliation probably made it feel existential on some level, a case of doing or dying on top of plain nationalism. To assert their independence, because an empire by definition can not be subordinate to any other power, Korea went through modernization reforms and proclaimed itself an empire in 1897 despite patently not being an empire (the justification being that Korea had historically once been divided into three kingdoms), but it was too little too late and IJ annexed them just a decade or so later, which absurdly must have seemed to justify IJ's worldview that one was either a colonizer or the colonized.

There was a political cartoon about the Boxer Rebellion where the Eight-Nation Alliance is seen carving up a pie labeled "China" as a Chinese man looks on in outrage, and it's so fascinating, because the Chinese and Japanese man (who, side note, appears to be the only genericized ENA rep rather than a caricature of a specific person) are drawn in the same racist way yet Japan was nevertheless by the turn of the century seen by the West as being at least close to same level - the full understanding of how far Japan had come on the world stage would arrive just a few years later when Japan kicked Russia's ass in the Russo-Japanese War.

The Russo-Japanese War is particularly interesting in regards to the disparity between how IJ viewed Asia and the West, because one thing that particularly shocked the latter was how 'gentlemanly' IJ's wartime behavior was, treating prisoners and civilians with boundless (and most importantly for the IJ, vocal) generosity. Yet, just a decade earlier during the First Sino-Japanese War, they committed many atrocities that would foreshadow the many they committed in the second. Unsurprisingly, the West quickly forgot the PR scandal IJ made for itself with the Port Arthur massacre of 1894 and declared that it's conduct in the war with Russia (keep in mind I am paraphrasing a turn-of-the-century opinion in a historical context!) "proved Japan is civilized".

It's interesting to think that the person who sent Perry to open up Japan and had this huge radical affect in causing Japan to become obsessed with modernization and catching up, which led to as much as it did, was...Millard Fillmore. Who Wikipedia notes is ranked by historians as both one of the worst and least-memorable presidents ever. The fact that he's in a very real way responsible for one of the most evil regimes in history is probably underdiscussed.

[owlsong74]

Incredible how much WDW keeps looking less and less attractive to visit as a west coaster and theme park fan.

The two coasts continue to merge. This happens with specific attractions and areas opening on both coasts (Mickeys Runaway Railway, Star Wars Land), with attractions coming nearer each other (hatbox ghost in WDW's Haunted Mansion), and in overall theming: Disney seems to have given up on park-specific theming, instead focusing on putting in movie based attractions wherever they can best excuse them, on both coasts. So even if the new attractions aren't exactly the same, the watering down of Epcot, Animal Kingdom, and DCA make them feel less and less different, and the former two much less of a draw.

Also, from the perspective of a fan of theme parks and their history, WDW is consistently watering down or removing its unique, historical attractions. The Great Movie Ride is gone, and replaced with a ride we have at DL now. Impressions de France plays second fiddle to a Beauty and the Beast singalong. The Country Bears will sing Disney songs instead of the country songs they, as characters, were literally designed around. The unique Dinosaur is getting replaced by an Indy ride that won't be starkly different than the ones that already exist. Maelstrom was a flawed ride which badly needed an update, but was far more interesting and had more character than the Frozen attraction there now (which is also getting copied into Hong Kong and Paris parks). Illuminations got replaced with a character show only slightly more Epcotish than an MK fireworks show (and was bad to boot, tho the quality is less relevant to my point). Even the hotels are getting genericized.

Finally, WDW has gotten significantly more expensive while customer experience has gone down. This is happening on both coasts but affects the decision to make a long, far away trip more than a local one.

There are still things that draw me to WDW. I still prefer their versions of several attractions found on both coasts. World Showcase and Africa and Asia in AK are still gems. Spaceship Earth needs a top to bottom refurbishment but is still one of my favorite attractions period. But its harder and harder to justify the expense and time of a trip, in comparison with just visiting my local Disneyland again.

My issues with how Disney is using their parks these days affect both coasts, but its striking to me that these changes make a cross-country trip more unappealing, in addition to everything else. I wonder if the same holds true for WDW folks and visiting Disneyland.

[fannishstuff]

"I listen to anything but ___" There is genuinely absolutely zero chance that your music taste is bizarre and your experience broad enough to include "everything" but you haven't found a single enjoyable piece of music under the label "rap." Do you enjoy the albums of elderly farmers singing out of tune folk songs that the Smithsonian collected for archival purposes? Do you know the different genres of Chinese opera and have an opinion on your favorite? Did you think the "Catholic psychadelic synth folk" above was a haha funny joke genre?

It's like you're a poser but for genericism. If you "like all music" you'd better be able to back that up.

"i mainly listen to pop-punk and emo and alt rock" yay! yippee :)

"i listen to everything but rap" no!!! first of all this information is so much less useful than just hearing what you do like. second of all, no the hell you do not listen to Catholic psychedelic synth folk. you're just singling out a distinctly black genre for Some Mysterious Reason

[turtlemagnum]

look, i had as much fun with AI dungeon as any red blooded gamer did during 2020. but the appeal wasn't the quality of the writing, the appeal was the Everything Gets Thrown At The Wall, Nothing Sticks aspect of it. it was goofy, random, and speaking as a writer it was Not good writing. i like to write and play fantasy stories, and one of the biggest flaws of AI dungeon was how set in its ways it tended to be. i'd try to write thoughtful, intelligent orcs and it always course corrected to the thoughtless, violent orcs of the kinds of fantasy i'm not a fan of and would often even use the word "orc" interchangeably with "ogre" and "troll". i'd try to write a lower class, downtrodden and inelegant elf and the AI would insist on her otherworldly beauty and grace. i'd write in a race that was exclusively female and a race of genderless fish people, and in both cases the AI defaulted to characters of these races being men. and it's because generative AI isn't actually writing anything, it's just an aggregate of a bunch of already collected works, and as such tended towards genericism. orcs are most commonly written as thoughtless beasts, so the AI couldn't even wrap its head around an orc that was as intelligent as any person was. elves are supposed to be elegant and beautiful, so the AI couldn't even fathom an elf that existed like a normal person. and since most stories are about men, obviously these races that flew in the face of that were as incomprehensible as an eldritch god.

now, i'm not one to say that AI art isn't art. anything can be art, and i feel like every group that's proclaimed something to be "Not Art" has been on the wrong side of history. and to be blunt, i'm not especially interested in debating what is or isn't art. but AI art, and particularly AI writing as that's my area of understanding, in my opinion isnt good art, and i think that's more important. i dont exactly understand the technology on an intimate level but from what i do know it's just a byproduct of the fact that it's always regurgitating an aggregate of what most writing is, and as such tends towards genericism. which isnt to say that works that are derivative are in and of themselves innately bad, it's just that in my experience AI writing is also just not good writing on a practical, prosaic level either, meanwhile even a derivative story if written by a competent writer can be executed in an interesting way.

[emotedllama]

Well. I watched the WondLa show yesterday and it was bad. I didn't end up minding the visual style that much (although I don't think a single design was changed for the better), but the knockoff Disney tone was just awful. It totally genericized everything about the story.

So I guess my childhood fandom won't be reignited, oh well!

[jonathankatwhatever]

It’s now 25 Jan 2024, and I’m still not entirely comfortable with this keyboard, but it is getting better. I find it goes better if I rest my left elbow and keep the right suspended. This keeps the right from getting lazy and not reaching the arm to the keys. With the keyboard tilted the right, the left’s motion is easier. I wonder how many variations this will permute through.

I can now get into some really tight folds and can squat nicely below parallel again.

This was not what I had in mind when I started typing. Caught one: easy to tilt to the right, and that lowers my elbow an inch or two, and that makes the move toward the upper right, which is the delete key, easier. But that’s a lazy position because it makes that one move easier and the others worse. The promise of an undeveloped land versus the reality of the paths it took.

I’m in some scary stuff. Do our paths ever coincide? I have no idea what because I can’t impose a version as though I define how reality occurs. I saw these VR images and they show something interesting, which is the genericization of your image so it fits a blunt sexual fantasy, meaning it is not you but a generated image of viewpoints or perspectives, those multiple takes which are pairs, which all synthesize to very straightforward Attachment of a structure that says looks like this person to any sexual role. Like the I Dream of Jeanie concept in which the issue is the complexity of the story versus the cartoon quality of the fantasy, with that solved through this process, which is Recombinance, because that is how the elements recombine, which is where the associative comes in, meaning this provides a solution to the ‘what fits between you and the mirror’, meaning that if we take you and the mirror then what fits between is the construction which is the solution which fits. That’s how you build toward and away from ideals.

That is, because the structure is within a structure, the old container idea often used in portable or modular programming, so the structure itself is composed or constructed. I keep saying the same thing in slightly different words. Keep right elbow out or the delete key squeaks.

I was looking at Dedekind groups, which led to the quaternion group, which led to the cycle graph, which led to the understanding that this maps to IC and Triangular over gs. It also maps to the Irreducible, which in the elements of the Q8 group are represented as -1 and 1. That answers a minor technical issue I had with the idea, which is why they represent the identity element as they do, typically I see as an e with a bar over it to indicate that it’s the commuter because if you square the other terms, meaning i, j and k, then you get that e bar, which is -1. To be blunt, that means you can switch at -1. This is an elementary example of pairing, isn’t it?

I really and truly forget that I have no idea what happens next in my head. I had no idea this was coming, but it is of course the exact answer needed and the answer which fits to the count of 2 perspective. Don’t have a name for that counting not of 0 to 1 but of 1 to 2, and thus from 2 to 1 rather than 1 to 0. That basic counting shift literally enables almost everything.

One reason is that pushes the count from 1 to 0 off a step, which then allows pairing across that gap to whatever Ends one sees or which occurs if you’re being fatalistic. Remarkably clear insight. This is how Joana thinks, and all I’m doing is accepting that this

I need a break. The illness situation here is not good.

[punkitt-is-here]

It's fascinating to me to encounter people on the internet who choose explicitly not to say what their pronouns are.

To be clear, I respect your decision and applaud the firmness of your stance. I think that overall there should be less personal information on the internet and restricting pronouns is simply another step on the way to taking back internet privacy. The thing is, I have spent enough time, effort, money, and grief on the project that is myself that I am having difficulty grokking the desire not to have one's gender be known at all. I am totally disarmed by the attitude, and I attribute this entirely to my own experience growing up on the internet. I can no longer abide resigning myself to the perceived genericized maleness of the average internet user. I deliberately make my pronouns available I go on the internet, and I correct people in conversation when appropriate.

This is because, in my past experience (middle 2000s ongoing), if one did not have clear gender they got the masculine pronouns. Assigned Male By Forum Users wasn't just the default, it was the practice, the de facto, the un-inspected habit of the english speaking internet user. There Are No Girls On The Internet was then and to a degree still is a self-fulfilling prophecy. Even though it was obviously never the case, and that women have always been on the internet, sometimes it is easier to just go along with the flow to avoid the inevitable misogyny.

Keep it up, you funky little internet person. I like your horse comix! They are good.

Haha, I totally get it! When you work so hard on yourself, you want to make sure people see it right. I'm all for pronouns being readily available and stuff.

That being said, I like keeping em a secret for a couple reasons! They'll come out eventually, obviously, I'm a film student with a job and eventually someone from online is gonna meet me in-person and I'll be happy to let them know then and when I (IDEALLY) have a career that'll be public knowledge. I've got a perfectly good gender I'm very happy and comfortable with. For now, tho, since I've seen it a million times with artists online that i really love, I've found out that people tend to weaponize your gender/pronouns a lot. No matter what they are, too! I think I just get less weird comments and mail by just keepin' that shit a secret. Folks don't gotta know cuz it aint really that important. I like being able to speak online and not having one trillion internet strangers who think they know me because of some words in my bio cross-examine me every time I talk.

Also, it's fun! I've had people think I'm a cis girl, a trans guy, a trans girl, a cis guy, a nonbinary no-gender something or other, a neopronoun user, genderfluid...basically everything under the sun! And I think it's fun to see how people perceive me since I don't think you get that option a lot. It's honestly really fun to know that my online presence doesn't have any...like, gender coding to it, if that makes any sense? Because I've seen NO consistency in the guesses and I find that incredibly entertaining. Even some people who've known me for years passively online still use the incorrect pronouns cuz I've never clarified and I think that's just silly and fun. I'm glad to know that I'm very much not explicitly in any sort of "box" with the way I present myself, because I think gender is kinda silly in the first place.

So, uh, yeah! That's mostly my reasoning for not clarifying anywhere, hehe. :3c

[titleknown]

A Couple of Essays I Did On Artists' Rights And AI Art

So, I'm gonna be posting a couple of essays I did on Pillowfort on AI art and the talk about datasets, the second one having been written a couple of weeks after the first.

Long story short, while I'm sympathetic to traditional artists' concerns about the dataset issue, as a copyright minimalist I'm also scared about how much they're parroting copyright-maximalist rhetoric in a way that's taking some dangerous routes to be used as tools for big Copyright to further increase its death-grip on culture, and maybe we should use this occasion re-examine the way we think about copyright and how we handle the idea of the moral rights of the artist in light of this issue.

Long story long... well, that's why I'm putting it past the break:

On AI Art Datasets, Copyright and "Theft"

...On the AI art issue, I will confess as someone who's cared about how bloated and awful copyright law is for years, it genuinely disturbs me that everyone's talking about the dataset for... well, really most AI art as an act of deliberate theft, when the reality is significantly more complicated, and the conception of it as theft feels like it's pushing people down a dangerous path they're not entirely aware of.

The thing is, the LAION dataset, which everyone refers to, is basically an attempt to crawl the entire internet for basically everything that has alt-text and sort it by how well the alt text matches up with the images. That's it. It's literally like a search engine crawler with the intention of "map the internet" rather than "find art".

And it is very; very dumb and brute force. That's how it ended up going through medical records, not any sinister intent, but because it's so stupid it looks for anything public with alt-text.

You can and probably should take issue with that, because it indicates failures to anticipate this on LAION's part and massive security failures on the part of those holding the data (And really this whole issue is more about data privacy than copyright) but it's not an issue of their intent being bad; it's an issue of the failure to account for that factor.

Another thing, it is actually very bad at categorizing artists by name. Like, through my multiple checks in the "Have I Been Trained" site; unless you're a big name professional artist, it almost definitely will not catalogue you by name.

If someone's looking to use your name in a prompt to avoid commissioning you, they are fools swindling themselves, and you should disabuse those cretins of their illusions, if only to discourage that behavior.

That's not to say I haven't run into the works of smaller artists I know (Who I've informed when I have, for the record), it's just that for better or for worse their work is genericized...

....Tho if you run into any works from artists you know in a dataset search, you should probably tell them, just so they're informed and can have it removed if they want.

And, following from that, for clarity's sake, even if you are informed, I don't think it's invalid to be scared of the potential impacts of AI art and the issue of data privacy (Which issues with the dataset are, as Tangibletechromancy talks about). In fact, I did see a post on Pillowfort expressing those sentiments that was relatively understandable. 

And it's not like there's no issue with the way they draw from the commons. As I've talked about before on my Tumblr, it's abhorrent that certain models draw from the commons but then make their models closed source and put them behind a paywall. Because it's taking from the commons and giving nothing back. 

Hell, the only reason I'm able to find other artists' art and inform them so they can have it taken out of the dataset is because Stablediffusion at least makes its model Open Source and its dataset publicly visible, both of which I think should be required for anything that uses that public data scraping. 

And it sucks that Midjourney and OpenAI are getting a lot less scrutiny than Stablediffusion when SD is arguably the one doing it the most correctly (Though I have heard MJ plans to eventually make itself Open Source, which, we'll see), and the former two ought to be looked at with a lot more skepticism...

...But my point is, it's less of an issue of deliberate theft and more Google Maps finding that one house of Barbra Streisand's she didn't want found. It's a survey of the commons of the internet trying to get as wide a picture as possible, for a set of "rules" as to what images look like (Note the AI model stores none of the actual images) and the conception of that as "theft" is what disturbs me.

Because, the idea that that 1/600,000,000th of a random image might end up influencing another's work without authorization relies on the idea of any derivative works as theft. 

With that 1/600,000,000th it takes less direct inspiration than an artist doing a pastiche or; arguably; even drawing from common experience would; even accounting for the other factors influencing it in an artist's mind because; hey; the AI has that too; as this post from friend of the blog Tangibletechromancy talks about.

It's an alien form of such compared to humans, but it is a form of that, because the dataset is big and dumb and anyone who's worked with stuff like Stablediffusion knows it reflects how dumb that sort of learning is.

And criminalizing that would definitely have knock-on effects, as this post by Trent Troop points out. Disney doesn't want to outlaw AI art; it has enough of a treasure hoard of works which it owns the copyright to to train its own. It wants an AI only it can use while drawing from copyrighted works, while copyright walls off access to it to everyone else. 

And it disturbs me so, so much that a lot of people against this are fanartists not knowing the precedent this could set; or hell; even the fact that the guy who started the train rolling on this moral panic; RJ Palmer; got his start doing Pokemon fanart, because the legal precedent that criminalizes this could very easily criminalize that too.

And, on a personal level, I have heard that argument well before that people should "stop stealing and Be Original," by people who... basically want to criminalize the concept of derivative works in general. People who believe that copyright should be perpetual, despite the concept of copyright as anything other than a temporary legal protection being very; very young from a historical perspective.

Like, you would not believe the shit I have seen. I have seen my producing teacher in college, who's most prominent producing credit was one of the worst modern horror remakes, argue that copyright should be perpetual in a metaphor comparing art to a family gas station. I have seen one person argue that the concept of derivative works itself should be outlawed to prevent them from diluting the original author's intent.

That person ended up rallying most of the other folks on a Discord server against me and driving me off of there. Not that I'm bitter or anything.

And I see that pattern in AI art, the animating sentiment that "derivative works are theft," with the same old "But it's different this time" framework laid over it. And believe me, I have seen enough "It's different this time" sentiments to be deeply skeptical of "but it's different this time."

In fact, that is why I'm scared in a way that motivates me to post this, because upon seeing anti-AI-art arguments going on like in the comments of this one post by the Staff of Pillowfort, I'm like, "oh god, I've heard this before," and where I've heard it before ain't good.

I see people trying to make the treatment of AI art akin to the way the RIAA treats music, despite the fact that that would be a horrible idea as this post points out, and people talking about wanting Disney to "save them" from AI art even though; again; Disney's more likely goal would be to use their own in-house trained AI to cut jobs while preventing anyone else from using it.

And I am deeply demoralized by the fact that over the time I've cared about it, from a perspective of material change this issue of fighting back against the bloat of copyright maximalism has basically never gotten any better (beyond the "Luigi wins by doing nothing" concession of stuff finally being allowed to go into the public domain very slowly in the US), and is very likely to get even worse

I have been angry for years that there's been no real legislative efforts to; say; decrease copyright duration or expand fair use, and now I'm living to likely see fair use shrunk even more. I will curse RJ Palmer's name until the day I die for single-handedly sparking this moral panic and basically undoing decades of work by copyright minimalists to kill the copyright cop in people's heads over a matter of months.

I come to my positions on AI art from years of being angry about copyright bloat and seeing the same patterns in the idea of it as "theft" as I do on people who were defending our current copyright nightmare before this, and I wish more people would push back on that. 

And if you have concerns about AI art and want to shape it right, I will point out, we have a Discord server...

Thoughts on AI Art and Moral Rights

I had some Thoughts wrt the debate on moral rights with regards to AI art datasets that I figured I might as well share with y'all, because I think the issues raised are more complicated than a lot of people say, and not in the ways y'all might think.

Like, it's a common talking point in the pro-AI-art circles that, even if the fair use defense were cracked down on, big megacorps that own huge swaths of images; such as Disney, Warner, ect, could still use the images they legally own; without the permission of their creators; to train their own AIs.

Which could, of course, lead to the same nightmare job loss scenarios that folks are talking about; again using artists' works to replace them without their permission, except the tools are behind a corporate wall and with no copyright ambiguity because; again; they own the images wholesale.

I've in fact heard it argued that; with the whole attempts by anti-AI-art people to join with Big Copyright's astroturf organization to expand copyright, that's what Disney wants, more crackdowns on copyright so they can use their own AI and you can't.

So, it's not a case of respecting artists' rights or don't. It's a case of whether everyone gets to use this tech at full power, or only Disney/Warner/et al are able to use it while the public gets a significantly weaker version trained on Wikimedia et-al.

The artists' rights; as articulated by those who are against AI art; are already fucked either way. Which, I am not saying as a gotcha. 

Rather because, while I know which of those two options I'd prefer,I sympathize with the fact that it fucking suuuuuucks for those creators who care about the moral rights of artists, and I want to examine institutionally why things are like that.

To start, a question: Why; beyond the legal reasons; is it okay for monopolists like Disney to violate creators' rights to control their work in that way; but not for wider-scale open-source projects like StableDiffusion to?

Some would say that it's because the megacorps pay and ask them. But, those often also end up as theft far more egregious than image synthesis programs do.

We all know the way that Spotify's "royalties" pay only pennies to creators and most of the actual profit to Spotify and the record labels that own the music themselves. Some even predict that that's how a license for using one's images in AI would go, which I think should give you pause.

And, we all know those stories of Marvel artists and writers wasting away in poverty and disease in their old age while Disney makes billions of the MCU and doesn't give them a dime. Totally legally allowable, they did get paid a pittance, once, but the billons more they never saw a dime of makes the difference in money not given to artists between them and the AI's unauthorized use more or less academic if we're going by sheer proportion.

One could argue it was even worse in the long-run, because AI's use doesn't technically force the subject's art behind a copyright wall and prevent the original user from using it, whereas the copyright landlords do, but that's probably it's own debate I can't get into at the time.

My point is, what the megacorps do is just as much theft if not moreso than what image synthesis training does, and the thin veneer of payment only obfuscates the vast degree of theft they do, which they only get away with because of how thoroughly it has been normalized.

And that's even before we get into the fact that it's hard to say you "consented" to it when your choices were "have the thing not exist and starve on the street" or "give us total control over what you create/"

But then, the natural answer of course you'll probably say after that to my first question (Why is it okay if Disney trains on my work without consent but it isn't for SD) is, of course, that it isn't.

But then, if you think of it purely in terms of copyright law (ignoring the fair use arguments for datasets), there should be no problem with what Disney does. They were "given" the copyrights fair and square, in the same way you would "give" a mafioso protection, but it was still fully sanctioned by the copyright system.

And yet, in a moral sense, there is. And, I think an important idea to articulate why this is a problem is the idea of moral rights.

The concept of "moral rights" in art is one that I don't see talked about much directly. Long story short, it is the idea that the artist has the rights to not have their work mangled and to be credited. Notably, it is considered a separate right from copyrights, non fungible in the way those are.

Note also that, it does not legally exist as a concept in the US, at least not to the significant degrees it does in other nations. But, I've noticed that the way a lot of people talk about copyright basically conflates the two. 

Which makes me wonder how much this debate comes from a US-based perspective, but I digress.

Like, a lot of the dialogue I've heard on why copyright is sacred; and especially from those who think it should be perpetual, isn't just about economic fears, but about the fear of your work being messed with and warped by those who don't understand it. 

They see copyright as the end-all be-all when it comes to protections for moral rights, because of how interchangable the two concepts have been made in the public dialogue. Copyrights require authorization from the holder to work with, they place the mark of their creator upon them, therefore they are thought of as valid insurance of those.

But, the point I'm leading to is this viewpoint doesn't really work. Because copyright on its own is a godawful protector of moral rights.

Like, the problem with copyright as a protector of moral rights is, it depends on a landlord model of security, IE the idea that individual ownership of "property" (even intellectual property( rather than collective protections will keep you safe. 

The trouble is, as Cory Doctorow has pointed out with regards to regular landlords, not only does that sort of commodification create a grotesque incentive to make things harder for those who don't have it (Such as, say, small creators with new ideas crowded out by legacy IP), but in the end power always gets consolidated under that system under the big guys.

Lucas sold his creation to Disney, Eastman and Laird sold theirs to Viacom. The fungibility of copyright; the ability of it to be bought and sold on the market, makes it a terrible means of protecting moral rights if you have to sell it to make a living, because once it belongs to a megacorp, they can do whatever they want with it, and that consolidation makes it harder for artists like you to show up.

In the case of collaborative works done under big megacorps, it's even worse because you have to give away those rights from day one to allow it to even exist. Look at what happened to creators under the whole HBO Max purge, copyright did nothing to prevent their work from being erased.Copyright did not do a thing to protect their moral rights.

Even in the case of estates, Doctor Seuss would be rolling in his goddamn grave at the Ilumination Lorax, and I'm pretty sure you can faintly hear Tolkien clawing his way out of the earth at Rings of Power being made by Jeff "Sauruman" Bezos.Copyright did not protect their works from desecration.

But the way we conflate moral rights with copyright in the conversation is very useful for those IP hoarders who want to expand their grip over the collective creative commons. It creates a broad base of public support amongst working creators for these copyright power grabs even if, as Doctorow mentions, it only benefits the top players due to their ability to buy everyone out and use their monopoly power to squeeze smaller creators further.

In fact, to bring it full circle, that's why I talked so much about why we need to push back against the idea of "theft" in datasets. Because the rhetoric of "theft," only makes coherent sense through a copyright lends, because how different piracy or derivative works actually are from; say; physical theft. 

There's a reason why "You wouldn't download a car" is a widely mocked concept. And I think that the accusations of theft are doing that exact same work of conflating moral rights for the artists to control their work to copyright, and as I have stated before, that is a very dangerous game.

This is why, I think, those of us concerned about moral rights need to start imagining means of protecting moral rights beyond and in place of copyright, because the conflation of them with copyright not only is ineffective, but also leads to those massive power grabs that undermine moral rights via monopolies. 

And, when you think of moral rights beyond the lens of copyright, it opens up far; far more avenues of thought to you.

Like, for example on AI art, the idea I've heard to focus on image scraping as a data privacy issue and not as a copyright one, because as a friend said the copyright angle is the least concerning use of scraped data for machine learning TBH. 

Or, to encourage practices such as Are We Art Yet's rules of ethics, which I think is deeply useful as a framework for engagement, and any AI art community should adopt them or something similar...

...Tho, thinking about it, a lot of the fears with regards to AI art and moral rights relate to Pillowfort user osteophage talking about how Tumblr; and to an even greater extent other social media sites like Twitter I'd assert; undermines community and the process of building communal norms, and how a lot of the fear is regarding to the breaking of communal norms in ways that'd devastate the commission/small artist economy.

Though that's its own ramble, which a friend of mine has sorta-addressed, but which I do want to give my own two cents on in the future.

Point is, I think the issues raised with regards to rights via AI datasets raise much deeper; more long-term questions regarding the nature of moral rights and the way they've been co-opted by copyright monopolists, and I urge you to direct your thoughts to those questions and what answers you might have for them.

If only because it will allow you to act much more wisely upon this topic rather than being lead around by the nose by copyright monopolists on it...

[kenyatta]

A LOT OF UGLINESS accretes privately, in the form of household goods, which can make it hard to see — except on the first of the month. Today’s perma-class of renters moves more frequently than ever before (inevitably to smaller apartments), and on moving day the sidewalks are transformed into a rich bazaar of objects significant for ugliness studies. We stroll past discarded pottery from wild sip ’n’ spin nights; heaps of shrunken fast fashion from Shein; dead Strategist-approved houseplants; broken Wirecutter-approved humidifiers; an ergonomic gaming chair; endless Ikea BILLYs, MALMs, LACKs, SKUBBs, BARENs, SLOGGs, JUNQQs, and FGHSKISs. Perhaps this shelf is salvageable — ? No, just another mass of peeling veneer and squishy particleboard. On one stoop sits a package from a direct-to-consumer eyewear company, and we briefly fantasize about a pair of glasses that would illuminate, They Live–style, the precise number of children involved in manufacturing each of these trashed items, or maybe the acreage of Eastern European old-growth trees.

It occurs to us, strolling past a pair of broken BuzzFeed Shopping–approved AirPods, that the new ugliness has beset us from both above and below. Many of the aesthetic qualities pioneered by low-interest-rate-era construction — genericism, non-ornamentation, shoddy reproducibility — have trickled down into other realms, even as other principles, unleashed concurrently by Apple’s slick industrial-design hegemon, have trickled up. In the middle, all that is solid melts into sameness, such that smart home devices resemble the buildings they surveil, which in turn look like the computers on which they were algorithmically engineered, which resemble the desks on which they sit, which, like the sofas at the coworking space around the corner, put the mid in fake midcentury modern. And all of it is bound by the commandment of planned obsolescence, which decays buildings even as it turns phones into bricks.

from Why Is Everything So Ugly? nplusonemag

[maaarine]

A Natural History of Human Thinking (Michael Tomasello, 2014)

“Social norms are thus mutual expectations in the cultural common ground of the group that people behave in certain ways, where the mutual expectations are not just statistical but, rather, socially normative, as in you are expected to do your part (or else!). 

The force of the expectations derives from the fact that individuals who do not conform to our group’s way of doing things often create disruptions, which should not be tolerated, and indeed, if individuals behave too differently it signals that they are not one of us (or do not want to be one of us) and so cannot be trusted. 

Group-minded individuals thus view nonconformity in general as potentially harmful to group life in general. 

The result is that humans conform to social norms for instrumental reasons (to coordinate successfully), for prudential reasons (to avoid the group’s opprobrium), and in order to benefit of the group’s functioning since nonconformity potentially disrupts this functioning (a group-minded reason). (…)

Human group-mindedness thus reflects a profound shift in ways of both knowing and doing. 

Everything is genericized to fit anyone in the group in an agent-neutral manner, and this results in a kind of collective perspective on things, experienced as a sense of the “objectivity” of things, even those we have created. 

Thus is human joint intentionality “collectivized.””

[balioc]

A Taxonomy of Magic

This is a purely and relentlessly thematic/Doylist set of categories. 

The question is: What is the magic for, in this universe that was created to have magic?

Or, even better: What is nature of the fantasy that’s on display here?

Because it is, literally, fantasy.  It’s pretty much always someone’s secret desire.

(NOTE: “Magic” here is being used to mean “usually actual magic that is coded as such, but also, like, psionics and superhero powers and other kinds of Weird Unnatural Stuff that has been embedded in a fictional world.”)

(NOTE: These categories often commingle and intersect.  I am definitely not claiming that the boundaries between them are rigid.)

I. Magic as The Gun That Can Be Wielded Only By Nerds

Notable example: Dungeons & Dragons

Of all the magic-fantasies on offer, I think of this one as being the clearest and most distinctive.  It’s a power fantasy, in a very direct sense.  Specifically, it’s the fantasy that certain mental abilities or personality traits -- especially “raw intelligence” -- can translate directly into concrete power.  Being magical gives you the wherewithal to hold your own in base-level interpersonal dominance struggles. 

(D&D wizardry is “as a science nerd, I can use my brainpower to blast you in the face with lightning.”  Similarly, sorcery is “as a colorful weirdo, I can use my force of personality to blast you in the face with lightning,” and warlockry is “as a goth/emo kid, I can use my raw power of alienation to blast you in the face with lightning.”)   

You see this a lot in media centered on fighting, unsurprisingly, and it tends to focus on the combative applications and the pure destructive/coercive force of magic (even if magic is notionally capable of doing lots of different things).   It often presents magic specifically as a parallel alternative to brawn-based fighting power.  There’s often an unconscious/reflexive trope that the heights of magic look like “blowing things up real good” / “wizarding war.” 

II. Magic as The Numinous Hidden Glory of the World

Notable examples: Harry Potter, The Chronicles of Narnia, H.P. Lovecraft’s Dream Cycle

The point of magic, in this formulation, is that it is special.  It is intrinsically wondrous and marvelous.  Interacting with it puts you in a heightened-state-of-existence.  It is -- ultimately -- a metaphor for The Secret Unnameable Yearnings of Your Soul, the glorious jouissance that always seems just out of reach.

It doesn’t so much matter how the magic actually functions, or even what outcomes it produces.  The important thing is what magic is, which is...magical.

This is how you get works that are all about magic but seem entirely disinterested in questions like “what can you achieve with magic?,” “how does the presence of magic change the world?,” etc.  One of the major ways, anyway.

The Numinous Hidden Glory fantasy often revolves around an idea of the magic world, the other-place where everything is drenched in jouissance.  [Sometimes the magic world is another plane of existence, sometimes it’s a hidden society within the “real world,” doesn’t matter.]  The real point of magic, as it’s often presented, is being in that magic world; once you’re there, everything is awesome, even if the actual things you’re seeing and doing are ordinary-seeming or silly.  A magic school is worlds better than a regular school, because it’s magic, even if it’s got exactly the same tedium of classes and social drama that you know from the real world. 

Fantasies of this kind often feature a lot of lush memorable detail that doesn’t particularly cohere in any way.  It all just adds to the magic-ness. 

III. Magic as the Atavistic Anti-Civilizational Power

Notable examples: A Song of Ice and Fire, Godzilla

According to the terms of this fantasy, the point of magic is that it doesn’t make sense.  It doesn’t make sense within the logic of civilized human thought, anyway.  It is nature and chaos given concrete form; it is the thing that tears away at the systems that we, in our [Promethean nobility / overweening hubris], try to build. 

There’s not a baked-in value judgment here.  This kind of magic can be presented as good, bad, or some of both.  Same with civilization, for that matter.

It’s often presented as Old Myths and Folkways that have More Truth and Power Than Seems Reasonable.  Narratively, it often serves as a dramatized version of the failure of episteme, and of the kind of entropic decay that in real life can take centuries to devour empires and ideologies.

This kind of magic is almost always the province of savages, actual inhuman monsters, or (occasionally) the very downtrodden. 

(I think it is enormously telling that in A Song of Ice and Fire -- a series that is jammed full of exotic cults and ancient half-forgotten peoples, all of whom have magic that seems to work and beliefs that at least touch on mysterious truths -- only the Westerosi version of High Medieval Catholicism, the religion to which most of the people we see notionally adhere, is actually just a pack of empty lies.)  

IV. Magic as an Overstuffed Toybox

Notable examples: Naruto, JoJo’s Bizarre Adventure

Or, sometimes, we care about what magic actually does.  More than that -- sometimes we want to see magic doing really interesting things, and then other magic intersecting with it in ways that are even more interesting.

The fantasy here, in simplest terms, is “magic can achieve any arbitrary cool effect.”  There doesn’t tend to be an overarching system that explains how it’s all supposed to come together, or if there is, it tends to be kind of lame and hand-wavey -- a rigorous system of Magic Physics, delineating the limits of the possible, would get in the way of all the cool effects we want to show!

Once again, this shows up a lot in combat-heavy narratives.  Less with the genericized D&D-style “magic is a fist that can punch harder than your regular meat fist,” and more with people throwing weird and wacky powers at each other in order to show how those powers can be used creatively to overcome opposition.  Sometimes, instead of combat, you get magicians using their cool-effects magic to MacGuyver their way out of problems or even trying to resolve large-scale social problems.  Issues of magic usage within the narrative being “fair” or “unfair” or “cheesy” are important here in ways that they generally aren’t elsewhere, since the fantasy on offer comes close to being a game. 

(Ratfic often falls into this category.) 

V. Magic as Alternate-Universe Science

Notable examples: the Cosmere books

This covers most of what gets called “hard fantasy.”  The fantasy on offer is a pretty straightforward one -- “magic has actual rules, you can learn them, and once you’ve learned them you can make predictions and achieve outcomes.”  It’s puzzle-y in the way that the previous fantasy was game-y.  It’s often a superstimulus for the feeling of learning a system in the way that video game grinding is a superstimulus for the feeling of rewarding labor. 

The magic effects on offer tend to be less ridiculous and “broken” than toybox magic, because any logic you can use to achieve a ridiculous effect is going to influence the rest of the magic system, and special cases that aren’t grounded in sufficiently-compelling logic will ruin the fantasy. 

Not super common.

VI.  Magic as Psychology-Made-Real

Notable examples: Revolutionary Girl Utena, Persona

This kind of magic makes explicit, and diagetic, what is implicit and metatextual in most fantasy settings.  The magic is an outgrowth of thought, emotion, and belief.  Things have power in the world because they have power in your head.  The things that seem real in the deepest darkest parts of your mind are actually real. 

This is where you get inner demons manifested as actual demons (servile or hostile or anything in between), swords forged from literal hope, dungeons and labyrinths custom-tailored to reflect someone’s trauma, etc. 

The fantasy, of course, is that your inner drama matters. 

My personal favorite.

VII.  Magic as Pure Window Dressing

Notable examples: later Final Fantasy games, Warhammer 40K

This one is weird; it doesn’t really make sense on its own, only metatextually.  I think of its prevalence as an indicator of the extent to which fantasy has become a cultural staple. 

The fantasy on offer in these works is that you are in a fantasy world that is filled with fantasy tropes.  And that’s it.

Because the important thing here is that the magic doesn’t really do anything at all, or at least, it doesn’t do anything that non-magic can’t do equally well.  It doesn’t even serve as an indication that Things are Special, because as presented in-setting, magic isn’t Special.  Being a wizard is just a job, like being a baker or a tailor or something -- or, usually, like being a soldier, because the magic on offer is usually a very-simple kind of combat magic.  And unlike in D&D, it’s not like magic is used only or chiefly by a particularly noteworthy kind of person.  It’s just...there. 

The great stories of the world, in these works, don’t tend to feature magic as anything more than a minor element.  The point is to reassure the audience that this is the kind of world, the kind of story, that has magic. 

-------------------------------------------------

Thoughts?  Critiques?  Other categories to suggest? 

[magnetictapedatastorage]

seriously the way EVERYTHING is a genericized brand name though. bandaid. granola. saran. its endless

[eiirisworkshop]

Soul

2020, Dir. Pete Docter, Kemp Powers

Overall Quality ⭐️⭐️⭐️⭐️⭐️5/5

Entertainment Value ⭐️⭐️⭐️⭐️⭐️5/5

Story ⭐️⭐️⭐️⭐️⭐️5/5

Visuals and Craft ⭐️⭐️⭐️⭐️⭐️5/5

Soul is a beautiful movie, both in terms of visual artistry and in what it shows and says about people, about passion and fear, about life.  Pete Docter—who is also the mind behind Inside Out, Up, and WALL-E—clearly has a feel for the depth and nuance of human emotion, and it is my understanding that we have Kemp Powers largely to thank for taking Docter's initial story of a soul reluctant to experience life and turning it into the celebration of Black American community that the finished film is.

The story is one of learning to appreciate life, to live for the sake of the little things that bring you happiness whatever they may be, to not tie yourself up too much in the idea of having to have a particular goal that you must constantly strive for and achieve otherwise you're a failure.  The film tells its characters—and through them, us, the audience—not to be so hard on yourself, and also to be gentle with and mindful of each other because we all have struggles and we all impact each other.  It's very astute in its assessment of human behavior and the message is one that I think we all could use to take to heart.  That's one thing that Docter's films all seem to have in common: they teach, directly without being heavy handed, lessons about life that are just as important for adults in the audience to learn as for the kids.

This message is taught through interpersonal relationships (with only the slightest passing mention of romance!) that center the importance of community.  School, the way a good teacher can impact kids, the cultural importance of the barber shop and of matriarchs in the Black community are all given particular emphasis.  There's a groundedness to the world of the film—highlighted, not detracted from, by the fantastical elements—that's artful, and often subtle. The subway is grimy, but the music being played by a busker in the station touches the heart.  The neighborhood with its driveways full of taxicabs is, by that wordless detail, shown to be working class, and it is precious and it's home.

There is a sequence that does deserve the warnings floating around about it, because it very accurately portrays the experience of a sensory overload and may be triggering for some viewers, especially because it comes on abruptly, but it's brief and only happens once. A later sequence deserves a similar warning for portraying the experience of a rejection-sensitive dysphoria or similar anxiety spiral.

The visuals of the film are stunning.  The etherial texturing of everything in The Great Before is reminiscent of the soft effervescence of Joy in Inside Out and feels appropriate to the setting.  The real world, living world sequences are where the animation really shines, though, particularly in lighting and the rendering of Black bodies—two things which are ultimately closely tied together.  

One of the trickiest things CG animation tackles when attempting photorealistic texturing (which the vast majority of CG films do, even when the designs are stylized) is recreating the way light interacts with human skin.  The difficulty in getting CG skin to not look wrong and deeply unsettling is, for the record, exactly why the first Pixar feature film is about toys.  Plastic is relatively easy to render.  If anything, “looking like plastic” is the default mode of CG animation.

Anyway.  Skin is made up of layers of varying translucence, so light passes into it some before bouncing back.  That process behaves differently on lighter vs darker skin because dark skin isn't just darker, it's more opaque—more, but not entirely.  Now, in the early days of photorealistic CG we were actually better at dark skin than light—in Final Fantasy: The Spirits Within (a groundbreaking foray into CG filmmaking from 2001), the character who looks distinctly the least plastic is Ryan, the one Black man among the cast.  But as time has gone on, an enormous amount of time, money, and effort has been poured into attempting to perfect CG renderings of light skin, while dark skin has been neglected, so we've gotten white characters (and a few lightskinned characters of color) who look more and more human while the few Black CG characters out there have lagged behind in plastic land.

For Soul, the time, money, and effort has been put in to close that gap.  We get Black characters in sunlight, under stagelights, in dim oblique lamplight, under harsh yet innadaquate subway fluorescents, and they look human.  Their skin looks like Black skin—there are a couple moments in particular that really capture the kind of soft, refractive glow you get from dark skin in bright light, which I have never seen animated before.

The lighting team for this movie deserve all the accolades they can get.  In addition to the above, Soul features the distinct look of extremely high-gloss black piano finish under various lighting conditions, brass instruments of varying levels of polished under various lighting conditions, city streets at all times of day, a sequined dress under stagelights, sunlight filtering through the semi-translucent leaves of a maple tree in seed—all of which look exactly right—and a square of sunlight through a window that just looks warm.

The character designs are stylized in a way that reminds me of the work of caricature artists you often find in city parks—each individual's physical characteristics are emphasized such that they made even more distinctly them, which is pretty much the exact opposite of the genericized, stereotyped, frankly racist character design that's still all too common.

I have seen some not-entirely-undeserved critique comparing Soul to The Princess and the Frog for having the Black main character spend a significant portion of the film in the body of an animal.  I feel like there's an important difference between the two, though, in that the protagonists of Princess and the Frog are transformed into animals so that for the bulk of the movie there are not Black bodies on screen and the story is divorced from those characters' existence as people of color, whereas Soul is a body swap situation where the experience of the characters existing in a Black body, living life, and interacting with the community around them is key.

There's pages more that I could write, analyzing and gushing about this film, the way it treats jazz music and the historical importance of that art form in Black culture and wider American culture, the impeccable rendering of various hair textures, the humor, the validation and value it gives to people who are often treated as lesser (public school teachers, children, sign-spinners), but for brevity let me just say:

This is an excellent movie.  It's worth watching.  It's good that it was made, and I think it may represent an important turning point in the film industry.  Time will tell what its legacy ends up being, but for now, it's a really enjoyable film.