This weekly roundup thread is intended for all culture war posts. 'Culture war' is vaguely defined, but it basically means controversial issues that fall along set tribal lines. Arguments over culture war issues generate a lot of heat and little light, and few deeply entrenched people ever change their minds. This thread is for voicing opinions and analyzing the state of the discussion while trying to optimize for light over heat.
Optimistically, we think that engaging with people you disagree with is worth your time, and so is being nice! Pessimistically, there are many dynamics that can lead discussions on Culture War topics to become unproductive. There's a human tendency to divide along tribal lines, praising your ingroup and vilifying your outgroup - and if you think you find it easy to criticize your ingroup, then it may be that your outgroup is not who you think it is. Extremists with opposing positions can feed off each other, highlighting each other's worst points to justify their own angry rhetoric, which becomes in turn a new example of bad behavior for the other side to highlight.
We would like to avoid these negative dynamics. Accordingly, we ask that you do not use this thread for waging the Culture War. Examples of waging the Culture War:
-
Shaming.
-
Attempting to 'build consensus' or enforce ideological conformity.
-
Making sweeping generalizations to vilify a group you dislike.
-
Recruiting for a cause.
-
Posting links that could be summarized as 'Boo outgroup!' Basically, if your content is 'Can you believe what Those People did this week?' then you should either refrain from posting, or do some very patient work to contextualize and/or steel-man the relevant viewpoint.
In general, you should argue to understand, not to win. This thread is not territory to be claimed by one group or another; indeed, the aim is to have many different viewpoints represented here. Thus, we also ask that you follow some guidelines:
-
Speak plainly. Avoid sarcasm and mockery. When disagreeing with someone, state your objections explicitly.
-
Be as precise and charitable as you can. Don't paraphrase unflatteringly.
-
Don't imply that someone said something they did not say, even if you think it follows from what they said.
-
Write like everyone is reading and you want them to be included in the discussion.
On an ad hoc basis, the mods will try to compile a list of the best posts/comments from the previous week, posted in Quality Contribution threads and archived at /r/TheThread. You may nominate a comment for this list by clicking on 'report' at the bottom of the post and typing 'Actually a quality contribution' as the report reason.
Jump in the discussion.
No email address required.
Notes -
"Oi, Bruv, Can I See Your Porn Loisence"
I've made this joke a lot in relation to the serially-delayed and maybe abandoned UK age verification mandate, so it's probably worth talking about Louisiana doing it for real:
Text of law here: it requires porn sites -- or at least most sites with more than 33% of their content matching the law's definition of 'harmful to minors', based around a modified Miller obscenity test -- to use 'reasonable age verification methods', or be liable for unspecified damages and court costs.
In its defense, the law does prohibit age verification companies from retaining "identifying information" (with again unspecified civil damages) after allowing access. For a variety of technical reasons such enforcement would be incredibly impractical, though: it's not clear how a person would find out, would find out who actually did it, and then show damages, without getting any complaint dismissed to early for meaningful discovery, or even with enough certainty to justify starting a lawsuit.
There's a pretty clear and wide potential for harm. Yes, yes, getting your porn tastes revealed to the general world isn't as embarrassing as it might once have been, even as the potential for Implications remains. And while I might trust the average incredibly-sketchy-porn-site or age verification company to secure my personal data that they're totally not supposed to be storing better than, say, Home Depot or EquiFax, that's kinda damning with faint praise. For someone that wants to host material -- increasingly, a necessity to speak in any meaningful sense -- this is a pretty tremendous landmine: not only do I get to wager what a Louisiana court might consider prurient or how it might do math, or what the risks of a teenager even finding my material might be, but also such fun imponderables such as "what impact might an unsuccessful lawsuit have on my job or position in the community".
It's... also not clear how this is going to work, at a pretty fundamental level. There are some deep constitutional questions regarding compelled and anonymous speech, and some annoying legalistic ones like the dormant commerce clause, and this is the sort of thing that's had SCOTUS involved before. And then there's annoying problems like grammar issues, whether the exceptions meant for exclude CDNs or avoid supremacy clause problems with CDA230 would also exclude booru or tube-style sites that do not create content, or how ads get handled period. Nevermind how much of a clusterfuck that "33%" threshold is going to be for all but the most overtly and specifically porn-focused sites: do courts have the infrastructure to handle this when even specialty sites can have millions of files in content? What happens if it changes, and how quickly does a site need to track changes? If a site decides to host a million pages of lorem ipsum or an old copy of wikipedia to pad their SFW side?
((Example: e621 has 3.3 million uploaded images, with
26% of them "Safe" and20% "Questionable" ratings, though this goes by different definitions than what the law here would involve, or even what non-furries would necessarily define them to be. Do I want to make bets on how the law would go there? No, because the answer is 'don't get in an incredibly humiliating interstate civil suit if you can avoid it'.))Some is just that none of the authors of the law nor the people promoting it can agree on what, exactly, the harms or scales of damages are. Peter Gheil points to Aella as the prototype of the 'who-cares' side of the progressive and libertarian perspective, and there's a lot of Culture War in that position existing, but there's a lot of positions outside of it (sfw meme). At the other end, there's people who want the extremely unsexy nudity excised from Maus, or object to Gender Queer over one comic panel out of hundreds of pages having portraying someone performing 'oral sex' on a dildo. Presumably Heinlein's later works fall somewhere in this spectrum, or outside of it.
But there's a slightly awkward situation where, in addition to the Baptist-and-Bootlegger coalitions, there's a separate compromise where this sort of law (Utah is considering a similar one, and California's regulatory apparatus might accidentally invent it by parallel means) is vague enough to marry people who simply don't want their ten-year-olds stumbling across the weirdest porn possible after typoing a web search, those who think a seventeen-year-old seeing a nipple will immediately and irrevocably twist his or her sexual orientation, the TradCaths who think showing ankles can lead someone down the path of temptation, the feminists that think showing PIV or bondage will push men to rape or domestic violence, and the feminists that think maybe sadomasochism should start in the late teens, along with every possible or plausible position in-between. Actual policy implementations are going to get a little rougher when practice comes about.
On that bootlegger side, some sites have voluntarily complied: MindGeek-related sites (such as PornHub) have begun requesting Louisiana clients to provide driver's licenses to the third-party LAWallet (which is its own weird mess). MindGeek had been an early adopter for that currently-mothballed UK version and has done some technology work on the verification side, along with being a pretty high-profile target, so it's not a huge surprise, though in turn it's far from clear how many other companies would want to work with them. Or comply at all.
In turn, though, it's hard to not think about where this might go down the road. Many of the objections to porn here generalize beyond it, even if a number of the advocates of restrictions don't (currently) want to expand them. China has recently pushed 'video game addiction' as a concept to the point of restricting gameplay hours, and a general 'social media addiction' is a pretty common political talking point (and tbf, may not even be wrong), and there's been an increasing (and tbf, not even wrong) push to talk about how the human brain doesn't really finish maturing until whatever age the immature-brained speaker wants a matter to add restrictions to.
And a tool to bring identity to a wide swath of internet activity is a pretty nice weapon to leave around waiting for someone to be tempted by it.
A wide swath of internet activity we could really do without entirely without suffering any harmful effects whatsoever, in all likelihood.
People and teens especially had more sex when porn wasn't really available that much.
People like looking at porn. We can quibble about the meaning of harm, but "People's desires going less fulfilled" is at least some kind of downside, and I think "without any harmful effects" overreaches. The absence is the harmful effect.
-People got along fine without it. Also, the ban would only affect digital porn, not physical or other hard to copy porn.
-it has potential of ruining the lives of 10-30% of population due to addictive behaviors.
Are you seriously claiming 60-80% of people having slightly more enjoyable masturbation sessions twice a week is worth ruining the lives of cca 10% of population ? No, really, you are saying that ?
-if you think porn is bad now, what do you think it's going to be like when people will be able to get whatever sexual fantasy they have generated within a few hours in high resolution by NNs ?
I'm exceptionally skeptical that 10% of the population are 'ruining' their lives over porn. Even groups promoting the concept give closer to ~6% as the upper edge for the entire class of porn addiction (not endorsed), which in turn is more built around self-identification than serious personal impact, which in turn isn't the same as actually going from correlation to causation, nevermind all the way to 'ruining their lives'. More mainstream analysis gives significantly smaller numbers.
((And, uh, 'motions to the entirety of alcoholism discourse, or football stuff' for the sort of tradeoffs we're demonstrably willing to make as a culture.))
I'm pretty skeptical.
More critically, I think this points to broader problems. You don't have a toggle of 'people masturbate/don't'. You're openly talking about situations that would require limiting access to pretty generic ML tools, and that's honestly just the starting point. Even fairly 'limited' laws like this will discourage a lot of not-solely-porn speech, and anything broad enough to seriously slow porn 'usage' will unavoidably touch on broader works.
A serious amount of addicts are completely in denial about being addicted.
Number of people 'self identifying' is almost totally irrelevant. In addition, you have religious nutters who think they're addicted for no reason at all.
Look at behaviors - e.g. the number of 30-40 yr old Japanese who are virgins and have never been in a relationship.
Being skeptical because of samey still images now, when there's no reason to think NNs won't eventually be able to translate narratives into film or 3d environments is kind of weird.
What ? No, I'm saying that tools that'd enable people to create endless sexual fantasy by prompting are going to be possibly even worse than just kids having access to endless porn.
These won't be basic tools, but probably something customized and specialised.
This is an interesting measure, and a growing one (at least where virginity = het virgin), but I don't think it gives the numbers you'd need -- the numbers only jumped six and four percent for women and men respectively from 1987 to 2015. I've not seen good information from the 2020 survey, but it doesn't look from google translate like a dramatic increase.
((And that six-and-four percent is from every cause, when there's a pretty wide number of separate social issues in Japan pretty strongly discouraging interaction between the sexes that's a far more plausible cause.))
Oh, I absolutely think it will translate to new spheres, and to some extent already is in the process of doing so. But at least for furries, film and 3d environments are not especially prevalent even today (and those that do exist often suck: there's a few dozen VR furry projects, each more forgettable than the last), and the overwhelming majority of that e621 scrape would have been mostly flat media, too. The bar for furry-interesting content is lower: the majority of interactions with the sexuality-side of the fandom are still images or short fiction already. But even if the quality can be more consistent and variety greater, it's not more interesting, and I think that problem will continue (and probably be augmented, due to less available and more samey material to train on) even as tools expand to different media forms.
This has historically been the failure mode for a lot of procedurally-generated or programatically-varied content in the fandom (and, to my knowledge, outside of it). You can pretty easily toggle species or background or orifice or
clothinglingerieaccessories (and there's a small industry of games like Lilith's Throne built around doing just that)... but there's a reason Lilith's Throne has a configuration system that borders on the obnoxious even to an overwhelmingly tech and nerdy playerbase, and even using it to the hilt can still run into some content you aren't really interested in. Yet it can as quickly become samey, not just in that there's a limit to the available human-written content that surrounds the procedurally-generated stuff, but even 'new' stuff repeats the same themes and the same phrases, as a rut of over-optimizing smut, and the more heavily you (have to) tune the configurations to your interest the faster that happens.What separates 'samey' pieces from novelty and interest is at least the potential for surprise. Now, it's porn. I don't want to overstate the artistic themes presented by the typical piece: someone invites a plumber, there's a joke about a lemon tree, money shot, yada yada. But there's a reason even tumblr adult gif fandom at its worst didn't turn into people just resplicing the same handful of images with slightly different subtitles. Tautologically, this could eventually be done through ML, but it's not clear how you'd do or define that, without going past the bounds of interested topics: either the consumer is taking a directorial role to some extent in the initial creation, or they get surprise cuckolding at best.
For an example, the furry porn-specialized fine-tunes of StableDiffusion took about 200GB of porn and a few days on commercially-accessible (if high-end) gaming hardware now, which could be fairly said to be outside of the realm of the typical user. But that's now, with people making random guesses and erring on the side of caution (and trying to make overinclusive sets so other people can use them). Specific concepts have been taught or identified from less than 100 source images, sometimes incredibly bizarre ones. ((And, conversely, a furry-non-porn-specialized fine-tune could still output racy images, mostly due to the limits of the SD2.x safety checker.)) I think the likely threshold is going to end up closer to a few gigs than to a few hundred. To the extent that a typical small porn stash can't be used to train a model, that's mostly because people don't tag downloaded images.
I don't think you can prevent pretty widespread access short of banning the entire models, or heavily restricting access to model training.
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link
More options
Context Copy link