this is maya!

risk it for the biscuit, baybee.

[kalichnikov]

“Terf is a slur used to silence us” dang bitch I wish it worked shut the fuck up

[besoyoba]

Billy Kid 13\07\24

[fuckyeahchinesefashion]

OP: Share beautiful and practical stitching techniques.(cr 随心意针线)

[fuckthisshit-ugh]

please read this even if you're not indian. especially if you're not indian.

you've seen the jokes. india is "dirty", indians are "everywhere", why don't they fix their own country. ok. let me tell you why, and what's happening right now while nobody's watching.

first, the history you skipped: britain ruled india for two centuries and drained it. economist utsa patnaik's research estimates around $45 trillion in today's money was taken. the india that got independence in 1947 was one of the poorest countries on earth. every "why is india like this" question starts there. you don't get to profit off a robbery and then mock the crime scene.

now, what's happening RIGHT NOW:

in may the question paper for NEET leaked. that's the national medical entrance exam. over 2.2 million students took it. cancelled. years of studying, families' savings poured into coaching, gone overnight because someone sold the paper. students died by suicide after. these were kids who wanted to be doctors.

and out of that came a protest movement. a gen z group called the cockroach janta party, started by a 30 year old named abhijeet dipke, got 22 million instagram followers within DAYS. they've been doing a sit-in at jantar mantar in delhi demanding education minister dharmendra pradhan resign.

and then there's sonam wangchuk. if you've watched 3 idiots, he's the real engineer the character was based on. built solar schools in ladakh. invented ice stupas so himalayan villages have water. spent his whole life on education. he's 59 and he's been on hunger strike for 17+ days now. reporters went to see him this week and he was too weak to talk. and the troll armies are calling him "anti national" and "foreign funded". a man starving himself for other people's children. that's the thing here, anyone who asks a question is suddenly a traitor.

it's not one incident, it's everywhere you look:

• infrastructure worth thousands of crores failing within months of the ribbon cutting. tunnels collapsing in regions where heavy rain is a guarantee, not a surprise.

• youth unemployment near 10% for ages 15 to 29, 13.6% in cities. in a country of 1.42 billion where more than half the population is young.

• only about the top 3% of indians earn above 10 lakh a year. meanwhile billionaire wealth keeps ballooning and the headlines celebrate it.

• manipur burned for over two years. an ethnic conflict, tens of thousands displaced, and the prime minister barely said the word manipur while it happened.

• journalists raided. critics' houses bulldozed. activists sitting in jail for years without trial under laws like UAPA. our press freedom ranking has cratered.

• even the petrol is a fight now, ethanol blending pushed through while drivers complain about their engines and get no straight answers.

and through all of it, the PM, who came to power on religious polarisation (americans, sound familiar?), travels abroad for photo ops and has not held one single open press conference in over a decade. not one.

what i'm asking:

if you're not indian: share this. our media mostly won't cover any of it because the big outlets are owned by the same billionaires the government serves. the only reason wangchuk's fast is getting any coverage is people refusing to look away.

if you ARE indian: speak up even if you're comfortable. especially if you're comfortable. join the protests. add to this, correct me, translate it. staying quiet is a choice when nothing in this post touches your life.

india is not "dirty". india was robbed, and it's being robbed again, from the inside this time. but it's also the country where a 59 year old will starve for strangers' kids and 22 million young people will rally behind a party literally named after cockroaches, because cockroaches survive everything.

we can still save her. but only loudly.

reblog. share. don't scroll past.

[aperfumedpearl]

Lily of the Valley Folding Fan, made in Austria in the late 1860s

[this-is-trans-joy]

I don’t know if this is an obvious take or a hot take, but I think people need to start re-framing feminism as the fight for body autonomy as opposed to whatever this second wave revival gender essentialist bullshit we have going on right now. Once you reframe it in this way, it’s easier to understand intersectionality and why cis women are not the only people who need feminism. The lack of body autonomy effects cis women, trans people, intersex people, disabled people, poc, homeless people, sex workers, etc. and your feminism needs to include and prioritise all of these groups of people (which will include men btw) because feminism is about autonomy, not about establishing a matriarchy. Body autonomy is the biggest threat to the patriarchy, both with reproductive rights, LGBTQ+ rights, and even the right to not be drafted into military services. Once body autonomy is established for everyone, the patriarchy no longer has a leg to stand on.

And body autonomy does include things that you don’t personally like either. I was prompted to write this post after a series of bad takes from progressives, but one of them was re-hashing the Sabrina Carpenter album cover drama with “I don’t think it’s conservative of me to think that the album cover is a bad look when we’ve seen images of women being abused in this way” because I do actually think you’ve failed to understand feminism by projecting your morals onto a woman who was consensually expressing her own autonomy just because she expressed it in a way that you didn’t like or that made you uncomfortable.

Body autonomy also means unhealthy choices. Body autonomy also means regret rates. Body autonomy also means freedom of sexuality. Body autonomy also means mutilation. If you believe body autonomy has limitations and exceptions, then your feminism is most likely surface level.

TERFs are some of the biggest opponents to body autonomy, and if you find yourself thinking “oh people can do whatever they want with their bodies as long as it doesn’t harm them or make others uncomfortable” then you are far more susceptible to TERF propaganda than you think.

[this-is-trans-joy]

ooooh the radfems are BIG mad about this one

[captain-price-returns]

The death of credible government is darkly hilarious in a cosmic sort of way. Obviously, the FDA is useless now. Therefore, we must place our trust in more credible institutions, such as Taco Bell, which has been far more transparent and proactive in handling this parasite business.

[diamond-dangeresque]

ah. this is real.

[miamaimania]

“Through Beaded Veils” by Marcin Górski

[hjarta-deactivated20260124]

in my room on a saturday evening talking to myself like this

[purr-in-ink]

[saishuumusou]

A li'l WIP from today, came up with an idea for a Summer Bready design.

I will now go shove some pizza down my face hole.

[fihas]

what people don’t realize about tumblr sexymen is a character being called one is a [bad] thing. the label was originally applied to characters that were lusted over by straight women who refused to engage with any other character who wasn’t a white twink. that’s why oncest exists. that’s why those bill cipher and wheatley gijinkas exist.

sans was a special case because people preferred his original design (short and not conventionally attractive) however the “refusal to engage with any other character” was very much still there like look at all of the AUs that were made specifically because of him.

the amount of mischaracterisation a character has to go through in order to truly count as a sexyman (and there’s about 20 sexymen proper) would cause most people to have a brain aneurysm like it sucks so bad.

times have drastically changed since the onceler, and the range of characters people hype up is far more diverse than it once was. the time period in which a character was most popular (between 2012 - 2018) is the most fundamental part in making them a sexyman.

this is why any sexyman poll made after the one that killed queen lizzy is completely null to me. 90% of the characters in them are just regular fan favourites like get this common real life dude out of there. if you could talk about liking a character with a coworker and not get viciously made fun of by them then that character is not a sexyman like that’s the full truth of the matter.

(this is also why anime boys like reigen and komaeda often count. hate for anything anime-based outside of the internet was especially egregious before 2020)

also. there is a huge distinction between “sexy man” and sexyman. this is why senshi does not count. Because he is genuinely sexy. My final message. Goodbye

[gifs-of-puppets]

The Muppet Show (1976-1981)

[teaboot]

I hate you Ozempic craze I hate you 'heroin chic' I hate you weight loss ads on public radio I hate Burn Fat Fast ads every thirty seconds I hate you I hate you I hate you

[teaboot]

I grew up before the term 'thigh gap' was invented I grew up before 'hip dip' was invented I was born before 'muffin top' was a thing before 'clean girl look' was a thing before 'glass skin' was a thing before razoring off peach fuzz was a thing and I'm so so so fucking tired of us inventing new concepts purely for the purpose of convincing people to hate their own bodies enough to buy products

[teaboot]

Last time Tuberculosis ran through the USA a small number of people got it on purpose to look skinny and waifish and delicate and used makeup to look flushed and bony and when the Victorians figured out tapeworms people would infect themselves on purpose to starve themselves smaller and women and now in the year of our lord 2026 there is a noticeable fraction of the USAmerican population genuinely thrilled about a treatment-resistant microbial parasite that makes you shit and vomit your brains out for a month because side effects include weight loss and STILL we talk about being skinny like it's the natural default setting for all healthy people as if it's a self-sustaining standard and not an imaginary goal that we are constantly constantly constantly beating ourselves with a whip to acheive

[frentique]

[chefpyro]

oh

[persistentmanlyagitation]

wait

[it-a]

i am tired and i will never be free

[ms-demeanor]

To have a computer reliably recognize a picture of a red-winged blackbird the computer must first imagine the universe.

[ms-demeanor]

I am having complicated and frustrating thoughts on AI again (as always).

[ms-demeanor]

I don't know if I've expressed this but part of the thing that has been making me bugfuck insane about discussions around AI image generation is knowing people who have worked with computer vision for decades.

[ms-demeanor]

I should fire up my 2005 macbook with CS2 installed on it and edit a photo entirely with "AI tools."

[ms-demeanor]

This is a 1980 computer science master's thesis by Ellen Hildreth on computer image recognition and creation. (Link downloads PDF)

The paper demonstrates the development, training, and testing of the Marr-Hildreth algorithm for edge detection in digital images.

[ms-demeanor]

Every time I'm gearing up for a good academic rant on this subject I further entrench my hatred of copyright.

[ms-demeanor]

Look. I understand that people are concerned about training models. I really do get that, that people have intense feelings about their photos or drawings or image being used to train AI. I even kind of get wanting to weaponize copyright against that because you don't know what else to do with those feelings.

I am currently building a multi-decade chain of papers on computer vision and image generation to have this discussion, and I would like to do so in a moderately calm manner.

Unfortunately if I want to cite a journal article from 1983 in a tumblr post, the copyright holders of this article about image restoration and pixel randomization want $248 per section of excerpted text (<500 words).

So the calm has now gone away.

[ms-demeanor]

Anyway. The flesh pit guy appears to be an asshole but I'm very frustrated seeing people react to photoshop compositing tools as an unethical use of AI tools that undermines the craft or artistry of a project.

This is a 2024 paper by one of the Adobe developers who worked on those compositing tools going over how the tools work. For the record, this model was trained on licensed and public domain images only and the tools are run on device, so the copyright and environmental complaints that people frequently make about AI don't apply to these specific tools.

That paper uses research from this 2015 paper on AI detection of composited images. That paper uses the ImageNet dataset.

ImageNet's dataset is a combination of images sourced from image searches starting in 2009 and description tags generated by Amazon Mechanical Turk workers. The images are not owned by ImageNet, but were scraped from various internet sources in the late 2000s.

That project uses algorithms and processes described in this 2007 paper on the utility of a general image database and image segmentation.

This 2000 paper on image segmentation developed some of the models used in the later paper, based on a collection of 1000 Corel stock images.

That segmentation was based on statistical models in this paper from 1994, which used a small training dataset collected by the researchers.

That 1994 paper made use of the model in this foundational paper from 1984 on predictive pixel selection for algorithm-based image restoration.

That paper helped to refine the boundary-finding methods used in this 1986 paper, which was an improvement on methods from the 1980 Hildreth paper.

Both Hildreth and Canny cite this 1971 memo from the MIT Artificial Intelligence lab, which describes the process of using a computer to find lines in an image.

Personally I like this note from a revised version of that 1971 memo that shows that we are still currently dealing with some of the same problems in computer vision that people were 50 years ago.

"This program has no idea what a reasonable line-drawing should look like when it represents an image of polyhedra. Instead it is very general and will find arbitrary line-drawings. Observing the particular way in which things sometimes go wrong, one quickly comes to the conclusion that higher-level understanding of the scene being analysed could greatly improve the line and vertex creating phase of this program. As thing stand now this understanding comes only after the line-finder has done its work."

Okay so what's my point?

My point is that there is a long chain of models, statistics, and research that, stretching back to the beginning of computer vision, has been centered around figuring out the likelihood that one pixel next to another pixel is black instead of white. The computers have never been very smart, they have never understood context, and the improvements we have made from the line finder in 1971 to "harmonize" in photoshop in 2026 is a very traceable chain of refining how the statistics are calculated.

They hallucinate extra elements, they don't know what shadows are.

Computers are still stupid, they just do math a lot faster than they did in the seventies.

Harmonize is apparently a new "AI" tool available in photoshop that is capable of matching lighting, texture, and other qualities in photos that are composited together. One of the things that the flesh pit guy is currently being dragged for (aside from really seeming to be a pretty tremendous asshole) is using this particular type of AI tool as a time saver on his ongoing art project.

I want to have a conversation about this, by which I mean I want to make some arguments about this.

I'm writing this specifically about the harmonize compositing tool in photoshop (and similar tools like upscaling). This is not about using grok or generating whole images or whole elements of images, this is not about whether or not the flesh pit guy is an asshole. I will grant that he is an asshole and I personally find whole images generated with AI tools like grok to just kind of look bad and be really boring.

Work with me on this, and let's accept the premise that adobe's Firefly is an ethically trained model (up to 5% of the images used to train firefly may be midjourney images that were licensed to adobe as stock images after being generated by midjourney, but adobe pays creators standard licensing fees to train data and does not train on client data), and that the tools are run on-device and do not consume any more resources than creating a 20-layer document in photoshop would.

(Also I acknowledge that training a model uses a lot of power and resources but creating a video game or an animated movie uses power at a much higher level than playing the game or watching the movie on your own machine; i've got to limit the scope somewhere so I can actually ask the question I've got)

Again, we are granting the following before you respond to the poll:

If the model was trained on ethical data and

If the tool runs locally and does not use more power than your computer uses and

The art the tool is used on is a personal, individual project and does not lead to poor treatment of workers

With that context I think those AI Tools

Are unethical and reduce the artistry of a work

Are unethical but do not reduce the artistry of a work

Are ethical but reduce the artistry of a work

Are ethical and do not reduce the artistry of a work

Voting ended onJul 20

[ms-demeanor]

Reblogging with corrected final question on poll.

[ms-demeanor]

Editing a photo with only the AI tools (predictive image processing tools developed from models trained on large imagesets) included as a part of the 2012 CS6 license (and the autofocus on my phone camera from 2020).

Direct from cellphone, Autofocus

Posterize (edge-finding like the tools in the original linked master’s thesis)

HDR Toning (photorealistic high contrast preset)

Selective Color Adjustment

Glowing Edges (again, another edge-finding tool)

Rendered lighting effects (spot, high intensity, high angle)

Color halftone

Gaussian Blur applied to the color halftone (similar to how AI upscaling works, and is based on the 1986 paper listed above)

Selective color replacement

Cutting backgrounds 100% with tools that use the AI included in my 2012 CS6 license

I think the last two images here and the edge-finding images do a really good job of demonstrating how AI/Machine Learning image processing tools are statistical tools.

The cut-out on the left was done hastily with a magic wand selection tool with a color selection threshold of 20. What that means is that you click on the image and the tool picks all of the pixels that are contiguous with the pixel you clicked on and are within the tolerance. If you broke down the image pixel by pixel you'd be able to get the exact count of what percentage of the image is what color, you could, in theory, remake the image as a barcode of just the proportions of each color using this kind of image analysis. The image is math, the computer sees it as math, the algorithms see it as math, the models see it as math.

The cut-out on the right demonstrates how the "AI computer vision" makes educated guesses (predictions) about the image. That was cut out using a magnetic lasso tool, which you guide around the vague shape of the image with your mouse. The program creates a selection line that follows color variations in the edges of the image that you're cutting out. On a very high contrast background (so if I was standing against a white wall) this can be very accurate. On a lower contrast background, there's more variation as the program guess wrong about which way to move the selection line.

The program doesn't see the photo as a picture of a person being cut out, it sees a collection of pixels of various colors and calculates what the probability is that a pixel of one color is likely to be next to a pixel of another color based on the the difference in color of the pixels.

There are very, very accurate selection tools in modern editions of creative suite - when I used the sky selection tool in my photography class in 2023 I was *stunned* by how accurate it was and how much time it saved. But all of that is just a more complicated form of the line finder in the 1971 memo that had trouble distinguishing between solids and shadows in areas of reduced contrast.