@gattsuru's banner p

gattsuru


				

				

				
10 followers   follows 0 users  
joined 2022 September 04 19:16:04 UTC
Verified Email

				

User ID: 94

gattsuru


				
				
				

				
10 followers   follows 0 users   joined 2022 September 04 19:16:04 UTC

					

No bio...


					

User ID: 94

Verified Email

I think that's useful to know, but I do remind everyone that a hacking group burned an ImageMagick zero-day on FurAffinity once. There's an upper limit to how far Rule of Induction brings you, here.

The virtuous man doesn't need to force himself to do the right thing, he does so out of habit.

"What is better? To be born good, or to overcome your evil nature through great effort?"

(a video game dragon lol)

Yeah, I'd second this. Microsoft has weird partial outages for the sole legal download source for the entire .NET ecosystem for three days, and a half-dozen twitterites and their own github was the only place to care.

I'll join the crowd picking at whether AR-15s as a class are status symbols or even largely purchased as status symbols. Individual sellers can be expensive or extremely expensive, but for the most part they're a fairly standard and fairly accessible centerfire rifle. Of course, when you see someone with a high-end reflex site and no cleaning kit, then 'status symbol' is one of the more charitable options.

((And a lot of this stuff is at least partly about deniable and even self-deniable 'stores of value'.))

Blue Tribe... depends a bit on the subculture. Honestly, more than the political alignment. I think Blue Tribers are a lot more likely than Red Tribers to focus on custom plastic or cloth trinkets, where Red Tribers might be more likely to spend on custom woodworking and large metalworking art, but to the extent that's even true, it's only true statistically and it's easy to come up with fairly simple outside causes.

For Anti-Evil Operations specifically, this should link to the most recent post removed by AEO that they've heavily references as a WTF moment, as at best a violation of the use-mention distinction and at worst actively counterproductive. bsbbtnh's post here is not the sort of thing I'd want to turn the forum into a long-lasting debate on, but in addition to the post's removal claims to have received a week-long suspensions.

Unfortunately, most older AEO activities look like the underlying post have fallen off the API that camas uses for indexing, or the full account was hit in ways that make the posts show up as deleted for camas purposes. The first four AEO actions were in response to a thread about a mass-shooter (I think the Dayton Ohio mass shooter?), which were significantly less controversial at the time, and seem to be in the first category. The oft-reference straight-of-wikipedia 'age of consent' list seems to be in the latter category.

Then there's the incredibly bad blocking implementation, that two years in Reddit still hadn't actually formalized those new 'advocacy of violence' rules that were supposed to be out in two weeks, that the mods were getting admin mails asking if they had any questions and then never responding, so on.

There's a lot of birthmark removals done for teenagers, often early teenagers. I think the current standard of care is 'anything larger than a quarter', and middle school ages are pretty typical. There's a lot of overlap between the technologies behind laser hair removal and birthmark pulsed laser dye procedures, and those usually target six months to one year. Some of this is about cancer risk, but then a lot of the trans emphasis is about health risks (eg, hysterectomy is strongly recommended for trans men due to elevated ovarian cancer risk from hormone therapy and probably from puberty blockers). And most of the limits on traditional cosmetic surgery for teenagers are focused on parental consent.

I think there's something of a parallel from social conservative perspectives, but I think they're more separating things not on the basis of surgery or cosmetic surgery as a class, but more because of the invasiveness of the procedure, the higher (although still low in absolute numbers) chances of complications, and lower opinions about the efficacy or even relevance of the procedures (and lack of trust in publicly-presented numbers for the efficacy).

Copied signatures are part of it (indeed, it's pretty trivial to end up getting stock watermarks out of StableDiffusion), but StableDiffusion at least does pretty clearly recognize individual artists and studios, and not just mainstream ones. It's not just that "studio ghibli" or "greg rutkowski" or "artgerm" drastically improves a lot of prompts: "hibbary" are definitely recognized keywords.

On the flip side, I'm not sure that this ban will actually block even moderately well-curated StableDiffusion txt2img results, never mind img2img or textual_inversion (or both) approaches with original bases, or where it's part of a toolchain rather than the sole step. Compare how the rules against tracing are almost entirely enforced against pretty obvious copycats, while drawovers are totally accepted.

On the other hand, I don't know that it's great to motivate people to strip any AI-specific hidden watermarks out (even if I hope no one's using FA or e621 for a general-purpose art AI). Which will be the immediate result even if none of the enforcement uses them.

On the gripping hand, I can understand if the genuine motivation were more immediate. It's pretty trivial to pump out sixty or a hundred varied images an hour, even with a multi-step AI toolchain and human curation. And there's only so much of that you can get before that's gonna have downsides, in ways that FurAffinity's (awful and dated) backend really isn't built to handle.

And while there's arguments that AI-generated art will make on-boarding into the sorts of collaboratively-purposed art that builds communities easier, FA's "Our goal is to support artists and their content" points to a more immediate concern. I don't think StableDiffusion's there even for the simplest cases (eg, single-character sfws, character sheets) yet, but it's believable that it could be close enough to impact the marginal cases in months rather than years. Whether AI-generated art has 'authenticity' or 'reflects the soul of the artist' may end up coming to entirely different results than questions about whether a community filled with AI-generated art becomes onanistic.

(If you'll excuse the puns.)

From a quick glance, Weasyl and e621 haven't taken the same approach (yet), and their underlying approaches are different enough that they may end up resolving the problem in ways other than direct bans on the media. Outside of the furry fandom, DeviantART hasn't blocked it, and enough artists have moved to Twitter that I don't think it'll be an issue.

If you agree on this much, then I have one question: which party ends up as the Party of AI?

Probably people outside of the United States (and probably outside of "The West", tbh).

Red Tribers don't care about 'artist rights' that much mostly because Red Triber exposure to Artist-as-a-title rather than artist-as-a-career is antagonistic at best, but there's no shortage of available outputs from AI that will trigger Red Tribe discomforts and no shortage of places where Red Triber frameworks for innovative ownership will be in conflict. Say what you will for the merits of doujinshi culture, but at least it's an ethos: the United States has 'solved' its copyright paradox largely by sticking its hands over its ears and its inventor's paradox by regulating away large parts of it.

I don't think AI-generated art is going to get hit as hard, but I think the general treatment of it's going to end up dropping it into a similar place, where it's theoretically available but practically fringe-even-among-the-fringe.

Google is useless for Reddit comments, especially in big threads. camas works much better for finding specific content.

And my broader point, that AI will continue to improve in capability with time, seems obviously and irrefutably true.

I'll give a caveat, here. AI will certainly get better within its existing capabilities and within some set of new capabilities, but there are probably at least some capabilities that will require changes in type rather than degree, or where requirements grow very quickly.

These examples are easier to talk about in the sense of text. GPT-3 is very good at human-like sentences, and GPT-4/5 will definitely be much better at that. It very likely handle math questions better. It more likely than not will still fail to rhyme well. It is also unlikely to hold context for 50k tokens (eg, a novel) in comparison to GPT-3's ~2k (ie, a long post), because the current implementations go badly quadratic. There are some interesting possible alternative approaches/fixes -- that Gwern link is as much about them as the problem -- but they are not trivial changes to design philosophies.

Well this, I'd assume, is because it can't have any way to know what 'rhyming' is in terms of the auditory noises we associate with words, because text doesn't convey that unless you already know the sounds of said words.

Unfortunately, it's a dumber problem than that. Neural nets can pick up a lot of very surprising things from their source data. StableDiffusion can pick up artists and connotations that aren't obvious from its input data, and GPT is starting to 'learn' some limited math despite not being taught what the underlying mathematical symbols are (albeit with some often-sharp limitations). GPT does actually have a near-encyclopedic knowledge of IPA pronunciation, and you can easily prompt it to rewrite whole sentences in phonetic pronunciation. And we're not talking a situation where these programs try to do something rhyme-like and fail, like match up words with large number of letter overlaps without understanding pronunciation. Indeed, one of the limited ways people have successfully gotten rhymes out of it have involved prompting it to explain the pronunciation first. (Though not that this runs into and very quickly fills up the available Attention.) Instead, GPT and GPT-like approaches struggle to rhyme even when trained on a corpus of poetry or limericks: the information is in the training data, it's just inaccessible at the scope the model is working at : either it does transparent copy or it doesn't get very close.

Gwern makes the credible argument that (at least part of) GPT's problem is that it works in fairly weird byte-pair encodings to avoid hitting some of those massively diminishing returns as early as had it been trained on phonetic or character-level minimum units, but at the cost of completely eliminating the ability to handle or even examine certain sub-encoding concepts. It's possible that we'll eventually get enough input data and parameters to just break these limits from an unintuitive angle, but the split from how we suspect human brains handle things may just mean that this scope of BPEs cause bad results in this field and a better work-around needs to be designed (at least where you need these concepts to be examined).

((Other tools using a similar tokenizer have similar constraints.))

The "more teenage pregnancies" has an archive here; that itself was just controversial, but the author also started linking hundred+ page PDF links with hardcore NSFW content that made the moderators have to deal with it and just generally making clear that they were either trolling at best, and... well, "turning teens into baby factories shortly after menarche" was the only reasonable read of their post, but it was way closer than I'd like to be anywhere near.

For the more general problem, I think there's definitely too much emphasis on later child-rearing in ways that are counterproductive both for those who can't or don't conceive at 30+, and even not great for those who do (both in health outcomes for the child, available energy, age-related issues for both parents). I think clearing up a lot of (charitably) miscommunication about those matters is probably going to be more effective.

(or, uh, institutional support for surrogacy on massive and probably also incredibly controversial scales.)

That said it's probably worth noting what a lot of the post-1960 curve is responding to. While a lot of US state laws date back to the Civil War era, enforcement and social norms largely treated those as tools for very limited sets of circumstances, rather than general rules. There was a lot more treatment of 16-year-olds as 'adults' for these purposes, and they stopped (and changed a variety of federal laws!), for a reason through the early 1970s. There's a few different causes, but one of the biggest was that most of the male half of such cases were not themselves teenagers, and that quite a lot of this turned out to be incredibly predatory. Anything that even risks bringing that sort of problem forward is going to (rightfully!) be a major landmine.

It'd be useful, and it's definitely part of the debate (and not one specific to trans- or LGBT-stuff!).

But I think there's also a few other concepts included and that are in many ways more primary parts of the discussion:

  • "Exist" as in being possible to recognize or know about, even under assumptions that they don't need to replicate themselves into the future. This is more overt for transmen, which were probably vastly undercounted in every statistical analysis for a decade, but the flip side of 'LGBT culture is trying to glom onto every crossdresser' is that Eddie Izzard has come out as genderfluid -- a lot of people who were part of the 'not-labeled' set demonstrably do want to go with these deeper categorizations when they become aware of them, some of which were not previously things that even had names.

  • "Exist" as in be able to go through society in a viable manner. The classical example here is the older WPATH SoC that required six months of lived experience before hormone therapy or hair removal: this wasn't physically impossible, but at best involved a bunch of really bad decisions and never being able to use a public restroom.

  • "Exist" as in being visible to other people. This is the other side to the 'you can be trans as long as you pass and I never have to hear about it anywhere'; not only are some people just never going to pass well enough to meet every critic's standards, but there's going to be at least some people who don't want to pass in every metric, either because their desired presentation isn't going to 'fit' (eg, transwoman who wants pants with pockets), or because they've grown to like things like Pride parades or obnoxious amounts of lipstick.

  • "Exist" as in be discussed in specific places where (they believe) other matters of similar level of complexity are discussed. This is the other side of the conversation about whether gender nonconformity is appropriate for middle-school age groups.

  • "Exist" as in live. This is somewhat based around overestimates of not-transitioned suicide rates and of transitioning and post-transition reductions in suicide rates (and sometimes overestimates of bias crimes), but you can still end up getting a giant pile of bodies with even more skeptical estimates.

but not very far to replace 80% of all fan art and furry commisions.

I'm not sure it's quite that high or that close. StableDiffusion is very good at making portraits or fullbodies of a single character with few accoutrements, for some species, but it struggles a lot with complex prompts or contextual clues and some other species, and while there's some ways that this will improve with additional training and data, there's others where it may reflect a technical limit in its underlying approach.

That doesn't mean it won't happen eventually. It doesn't even mean StableDiffusion can't be disruptive as-is -- I expect we'll find more and more Photoshop/SAI/so on plugins that use it as a texture- or brush-like tool to add detail and form to individual components of an image. It does some things even great artists struggle with: interpolating a character from different perspectives or in different media using textual_inversion is really magic!

It's not that it can't make a character sheet. It's not even that it might not have the token width to input a prompt for a character sheet. It's that it's not clear the current approach can allow it to have the necessary contextual framework necessary.

Of course, that might just mean one decade rather than a year.

But then…can I at least have my own awards convention so that I know which books from this year aren’t utter crap?

I'd be interested in just a variant of this awards convention that isn't crap.

You can do really interesting things with an aimless and actless protagonist -- historically, Gulliver's Travels or The Time Machine, but my favorite example is the excellent Kino's Journey series. Chambers flubs not because the Wayfarers series lacks a goal, but for the same reasons the (much less conventionally woke and much more conventionally 'plotted') The Wrong Stars does: there's just not enough tension. Not that it needs to be high-stakes: both stories are, in the same way that Dragon Ball Z is high-stakes. But they have less actual conflict between what characters want and what they're doing than a Sesame Street episode, fewer drawing questions than Haibane Renmai or the average litRPG.

"Bots of the Lost Ark" is stronger in that there's at least something there -- you don't know why any of this is happening or who these people are, and you kinda want to -- but the characters aren't coherent enough to feel like it's important or urgent rather than author fiat. The best I can say about "Where Oaken Hearts Do Gather" is that it's interestingly experimental and has a clever 'twist'? But in addition to the experiment sucking, the format just doesn't drive you to care about the gotcha until you're almost three-quarters of the way through, and the best it does for theme is a self-referential 'oh, but maybe themes are mixed' footnote.

Never Say You Can't Survive is... not science-fiction or fantasy, or even really fiction. It's half how-to-write, half self-help book. Which isn't the worst example of unrelated junk that's been put forward for Best Related Work, with some of this year's pieces edging on the onanistic. That's not just a matter of philosophical or political disagreement: “How Twitter can ruin a life” is closer to my views, but it's still very much a writing-about-a-real-world-news-about-sci-fi-writing thing rather than actually Related. But still a long-standing problem. And while I'm not the target audience for podcasts, this doesn't really impress.

But you could do some really fascinating stuff with these pieces, and with the exception of Never Say, it'd be a editing change, rather than a deep change of scope or theme. It just doesn't seem important, any more, in the same way that Tor's not really an editing service to the limited extent it once was.

((There are some Hugo Awards that were serviceable or even good. I don't think I'd have voted for Desolation Called Peace, but it's pretty enjoyable a read. I'd rather Fan Writer go to media writers rather than infrastructure ones, but the WorldCon voters as a whole have long-favored infrastructure and Buhlert has more than paid her dues on that matter. Dune both goes without saying and isn't another godsdamned Doctor Who episode. Lee Moyer is an amazing artist in general, and the small gods project showed that off a lot even if the actual works were incredibly shallow. I've got mixed feelings on Jemesin's writing for the same reasons I don't like Bojack Horseman, but I've heard Far Sector's not bad for a Green Lantern series.))

I'm also generally unhappy about the repeats. A Memory Called Empire and A Desolation Called Peace aren't awful, as much as the latter had a little too much overlap in one of its twists with Ender's Game. But especially good competition like Project Hail Mary or Black Water Sister, it feels at best like it's a symptom of block voting for the same authors to repeat.

The infantile POTUS in powers of the earth as well, whose name escapes me.

"Themba Johnson" (and her understudy, "Linda Haig"), because for whatever the merits of MorlockP's writing might be, subtly isn't one of them. Although 'infantile' probably isn't the right measure for her: the character's point is that she's much smarter than she seems, she just applies that to political ends rather than technical ones. Note that whenever she makes a numeric 'mistake', it's in ways that make much better sound bites than the truth. (I think this is meant to directly contrast with some of the spacer leadership: Javier makes a few similar mistakes, usually related to identifying people or places of origin; that the differences don't matter in physical senses but do show relatively lacking social skills is a theme.)

Although agreed she's more a Clinton expy than any sort of steelman. For sympathetic grounder characters in Powers, you'd probably be better-served by Restivo (who's an 'honorable' soldier, if compromised by his loyalty to his commander), or Matthew. For sympathetic women, there's a pretty wide variety of spacer ladies (and a couple sympathetic young women).

Even as a Clinton expy, she's not exactly evil for evil's sake. The UN and US (correctly!) sees spacers as huge physical threats, in addition to acting as a combination of brain drain and tax shelter, not to mention the unlocked AI that's been using half of the moon as a playpen. These are just drastically different values from those of the spacers, and of most readers.

But it still begs the question of why the slate has been dominated by women unless the people running the Hugos would argue that women are innately better at writing scifi, or if it's some form of restorative justice, just how long they want to keep it up.

There's a "the top-5% women are better than top-% men, whether from socialization or other cause, at sort of the coalition management the voting system runs on". Which I don't think is particularly palatable for Hugos, but it's not the most damning indictment.

Personally I don't believe it's all that needed on Reddit.

I think it could be unnecessary, and still happen.

Tier 4 (going to have to try and model this one): Sexual attraction is the ultimate form of acceptance/'passing.' The way we can best support/validate trans or gay children is to validate their attractiveness to the other sex...? Someone would have to explain this one to me as well.

I think the steelman for this is 'sexual attraction' isn't really a coherent category, and a lot of things that social conservatives put into the 'sexual attraction' bin don't actually seem like central examples of what people are actually objecting to, but rather parts of t2 that just have additional cognitive loading.

The internally-used example here is something like the Jessica Rabbit: a style of dress and presentation that's charged... but not actually doing anything. Putting on thick lipstick and a sparkly dress isn't playing hide-the-sausage more than Rabbit playing pattycake was; to the extent the former is sexual and the latter isn't, it's because we've assigned a whole lot of identifiers-for-being-female-socially as sexualized (probably by a mix of taboos and mode expectations?) . But these same things remain as identifiers-for-being-socially-female, separately, and it's pretty common for trans people to glom onto them in that role, in ways that can exist separately from the sexual attraction (although sometimes it doesn't!).

There's a plausible argument that we don't get appalled over the same stuff when done outside of this specific culture-war context. Letting a pre-teen (cis) girl dress in gaudy costumes and make a mess with lipstick might get you shunned, but it's not going to turn into national news, and if we're talking your own kids, probably not get CPS called on you. We don't pass out prison sentences to everyone who lets a kid use an insufficiently-filtered internet connection. At the extreme object level, the serious harms caused by seeing someone's dick through their panties got Ace Ventura a PG-13 rating, and I'm not sure it was actually about that; RuPaul's Drag Race usually nets a TV-14 for broadcast. Or for non-sexual drag, Eddie Izzard probably isn't appropriate for pre-teens, but that's more because of the cursing than the dress.

This is a broader problem for the L, G, and B spheres, too (as well as fandom): it's not uncommon to see people worry about whether Pride parades allow under-18s, which makes sense in the context of Folsom Street Fair... but most parades aren't that, and the complaint remains. The nearest Pride for my situation's most adult situation is the rampant alcoholism, but that's shared with the nearest Nascar event, too.

((This is further complicated because a lot of advocates from either direction aren't aware how limited their understanding is, even as they're motioning about limitations in understanding. Progressives point to various young-teen or pre-teen beauty contests, except these are also things that the vast majority of socons find appalling, as naraburns points out. Conservatives point to endless twerkfests... but it's not like these are some unheard-of thing in straight culture. "Penis inspection day" is and was a regular joke on reddit and tumblr in relationship to trans politics, derived from an older UrbanDictionary meme, but it was also not an uncommon thing for schools to have either full-time staff nurses or contracted doctors who'd perform physicals in bulk, including the turn-your-head-and-cough bit, although this has thankfully fallen out of favor.))

The opposing steelman against is that just because something varies by culture and time doesn't mean we have to accept it in this culture and time, and you have to draw lines in the sand somewhere, and that socons have (sometimes even honestly!) drawn lines well before these points in the past.

And when someone says "Drag Kids is sexualizing children" only to be met with "no, you're making it sexual, you right-wing pervert, we're just having silly fun"--it's maddening. Like, really? I'm supposed to believe that you're putting your kid in a leather thong for silly fun? Be serious. If that's not grooming, nothing is.

To be fair, I don't think this is an accurate summary of what's usually going on with all these events. The Boise 2022 Pride Guide doesn't have much information on what Drag Kids was actually going to revolve around, but a follow-up "Drag Story Hour" references Kenni The Doll, which... I'm not a fan of the makeup and not going to try to figure out what's going on under any clothing, but it's not looking like leather thongs or most adult entertainer 'wear' (there are two other entertainers, but I can't find any clear images or references to them).

I'm not saying that leather thongs (or too-sheer, or too-short, or otherwise too-revealing clothes) never happen! But whether or not it's a central case for drag as a class, I think a categorical opposition to the topic needs to handle cases like Eddie Izzard, who isn't appropriate for kids, but in a British comedic swearing sense, rather than anyone getting aroused by it, or find a meaningful way to separate them.

And I'm not sure the latter option is solvable, exactly. "I don't want to know, and don't want to need to know, about your underwear" is absolutely a reasonable norm, and one I share and support, and I think you can get a very wide section of the American populace behind. As the discussion becomes more about "this style of dress is inappropriate" or "this style of dance is inappropriate", though, I think that level of agreement gets a lot harder.

For a less culture-war-centric example, the furry fandom gets a lot of askance looks for people who fursuit in public, especially where kids might see. After all, people have sex in (something that kinda looks like) those! Well, ok, bringing a murrsuit body anywhere near public is one of the ways to reliably get nuked by everyone else in the fandom, but just because people haven't had sex in that particular suit doesn't make it less sexualized. For example, there are some (very nice!) fursuits with very thick thighs and incredibly fluffy tails and highly pronounced toe beans. Which absolutely can be fetishized in actually-sexual ways, but are also things people just think are cool.

((Though to extend the metaphor, this doesn't make it ethically mandatory to permit: both for practical reasons and for credit card processor ones, there's a lot of restrictions on under-18 fursuiting, many not explicitly written down.))

The fandom's largely managed to set and evolve some norms around here (don't use murrsuit bodies for anything public or mixed-use private, full suits should cover as much skin as viable, partial suits that don't cover all skin should be worn with fairly concealing clothing, UncleKage will murder ban you if you turn into a PR debacle), but note that this is the fandom. Normies who run into the issue don't just come up with a different answer; normies come up with different answers from each other.

And I don't think we have enough culture-wide communication to really build or even discuss normie-wide norms, anymore.

Possibly in response to this thread?

I'm going to reiterate my normal complaints about Campbellian (or Jungian) analysis; this can be kinda funny, but it's so broad and vague it makes astrology look well-founded.

I think there's a surprising amount of useful information, even beyond the QCs. ((Indeed, one of my complaints about the QC roundup is how mediocre a QC can be and still get awarded, while some very good posts in the same time frame either weren't nominated or didn't pass the cut, and I'm not just or even mostly talking about my own posts.))

I think there are benefits to making continuing an older interaction difficult, given what happens to fora without rules against necroing, but I'm very much against the Eternal Now.

And, on the flip side, I do think it's useful to notice when people are very confident but wrong, myself included. It's good to not have it be easily visible, but it's also helpful to notice when someone makes a big crux of their argument about a generalized position they've violated in the past, or base their knowledge on personal experience or 'friends' that they would require them to be The Most Interesting Person In The World.

The NYTimes piece suggests that Hasidic schools do have worse outcomes on the 12-year-old reading and math tests than most low-income public schools. Hasidic schools counter that their students do better in (voluntary) high-school-aged tests, though it's not clear that those are much more representative. A lot of the rest of the numbers and arguments seem to be pulled or supplied from this YAFFED report, although they're a single-issue non-profit focused on this issue.

That said, the breadth of the difference looks wonky enough that I'd like to see the actual measurements, especially since I can't find any mandatory testing for private schools in New York or NYC specifically, and that the numbers don't make sense -- YAFFED repeatedly highlights the breadth and importance of this topic, with tens of thousands of students effected, and... I can't see how that matches with a total of 12 boy's schools, especially given the Hasidic preference for tiny classrooms. A different YAFFED report looks like it found low scores in a NYC-specific subset, but the numbers there don't look to match either. That report points to NYSED, which seems to love 40+MB Access or Excel files (why?!), and doesn't seem to cover non-charter private schools anyway.

EDIT: actually, the extent that this doesn't seem to match the internals of the NYSED researcher files worries me a lot about its general accuracy.

In terms of broader results: Hasadic communities have high poverty and welfare usage, but low reported rates of violent crime (although there's some controversy about how much domestic violence is not reported).

To some extent, it does mean that.

But I've also had one of my own posts QC'd which I'd requested to be removed from the QC lineup, and there's other pieces like this that were QC'd but basically a news link with a bit of context. I've also got mixed feelings at best about this one.

It's not that they aren't insight porn; it's that whether QCs are about 'posts that are good for the community', 'posts that present unusual perspectives or expertise or viewpoints', good effort-posts, or just 'things we want more', regardless of the merits of them as standalone posts, they weren't that.

Now, I'm not filtering QCs, nor would anyone want me to be, nor am I writing a surfeit of outstanding posts that need to be elevated instead. I've reported a few QC submissions, but not many. It's probable not possible to consistently get the best results on whatever metric, no implemented system can be perfect, and the actual project is still worthwhile.

But it's also worthwhile to keep in mind its limitations.

That's true, though in turn it raises a few awkward questions. One of the possible explanations for a bunch of 'schools' that have very long classroom hours and culturally-encouraged out-of-classroom study, yet get these test results is that they're learning to read, write, and do math in another language.

((You can find the ELA and math questions online.))

Despite Juarez having 5 times the violent crime and homicide rate as Philadephia, its looks more functional and livable. The streets the cleaner, the infrastructure isn't as dilapidated, and there are more businesses. This is because the Cartels are very much running shit there.

... while I'll generally agree with the 'stationary bandits' thesis, this isn't a great example. Even ignoring the elephant in the room of "livable" with a high (and very unpredictable!) murder rate, the Juarez half of the video focuses on the city center and downtown : while even this area was unsafe back in 2007-2012, today it's more comparable to Philadelphia midtown or at worst North Central. The Philadelphia portion looks to be shots from one of the slums in North Philly, maybe Glenwood area.

Juarez's slums tend to be less obviously messy, but you can still find a lot of spots with trash, graffiti, and unemployed affiliated citizens everywhere, along with a lack of businesses. And the difference in quality of homes is pretty huge! Moreover, that difference right now is kinda a best-case scenario; in 2010, something like a third of businesses in the city either moved out or closed down, with corresponding problems.

More broadly, a lot of Juarez' infrastructure and business appearance reflects tremendous injections of outside cash, largely under the theory that the cartels were able to hire so many expendable and trigger-happy troops due to the rampant poverty, rather than some specific and intentional behavior of the cartel. While a few of the nice buildings and clean streets are specifically due to gang actions (if more in the sense of the gangs liking their own buildings and streets looking nice), a lot of the improvements have reflected state or foreign national investment.

((On the flip side, yes, a lot of the violence came about because of either cartel-on-cartel or cartel-on-government actions. Albeit not all of it; during the height of the 2010-era violence and again with some of the more recent smaller surge, there's been a lot of what's pretty likely 'personal' motivations.))

Some past conversation on this topic. I'll also point to the Times in terms of where gun control advocates want this to go next:

Creating the merchant code is only the beginning. Here’s what will need to happen next for it to help identify suspicious purchases:

  • Card networks like Mastercard and Visa need to not only adopt the code, but also enforce its use by merchants and payment processors.
  • Merchants must start using the code, and not obfuscate transactions by using other classifications.
  • Big retailers like Walmart and sporting goods stores — which themselves use different merchant codes — need to use the code at registers they use to ring up firearms.
  • Most crucially, the payments industry needs to develop and refine software algorithms for identifying suspicious activity based on the merchant codes.

And, more subtly :

Lawmakers like Senator Elizabeth Warren, Democrat of Massachusetts, publicly supported the plan...

Senator Warren, while better known for a hilarious incident involving a genetics testing kit, is also a member of the Senate Banking Committee, and chair of the Subcommittee on Economic Policy.