--- title: Counting the Countless: Why data science is a profound... description: An essay arguing that data science, as currently practiced, is structurally hostile to trans and queer existence because it demands quantification,... canonical: https://daily.dev/posts/counting-the-countless-why-data-science-is-a-profound-threat-for-queer-people-bsfcnbrhb twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Counting the Countless: Why data science is a profound threat for queer people | daily.dev og:description: An essay arguing that data science, as currently practiced, is structurally hostile to trans and queer existence because it demands quantification,... og:url: https://daily.dev/posts/counting-the-countless-why-data-science-is-a-profound-threat-for-queer-people-bsfcnbrhb og:image: https://api.daily.dev/og/posts/bsFcNbrHb.png og:image:alt: Counting the Countless: Why data science is a profound threat for queer people og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Counting the Countless: Why data science is a profound threat for queer people **[Lobsters](https://daily.dev/sources/lobsters)** · 15 min read · 0 upvotes · 0 comments ## Summary An essay arguing that data science, as currently practiced, is structurally hostile to trans and queer existence because it demands quantification, standardization, and ubiquitous surveillance that erase autonomy, contextuality, and fluidity. Drawing on Dean Spade's concept of 'administrative violence' and Anna Lauren Hoffmann's 'data violence,' the piece critiques examples like insurance companies tracking food purchases and Joy Buolamwini's gender-recognition bias research, arguing that reformist fixes (more diverse datasets, more inclusive facial recognition) only make surveillance systems more efficiently harmful rather than less harmful, since control and normalization are the point of these systems, not a bug. The author calls for building counterpower and refusing participation rather than trying to reform data science from within. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### Why did Joy Buolamwini's Gender Shades research recommend more diverse datasets for facial recognition, and what's the critique of that approach? Joy Buolamwini's 2018 intersectional analysis found facial-recognition and gender-recognition systems biased against dark-skinned, female-coded people, and her follow-up recommendation was to use more diverse training datasets. Critics argue this reformist fix ignores that these systems rest on an essentialist, biological view of gender that harms trans people regardless of accuracy, and that making facial recognition better at recognizing marginalized groups mainly sharpens law-enforcement surveillance tools already used against them. _Anyone weighing fairness fixes against surveillance risk in ML systems can track this debate on daily.dev._ ### What is Dean Spade's concept of administrative violence? Administrative violence, coined by legal scholar Dean Spade, describes how administrative systems like government IDs and legal processes force people into narrow, rigid gender categories in order to access basic needs such as banking, healthcare, or a driver's license. Spade argues that reforming these systems, such as adding more gender options to IDs, tends to legitimize state control over gender rather than removing the underlying coercion, and mainly benefits the least marginalized. _Developers building identity and data systems can follow this kind of ethics discussion on daily.dev._ ## Community take How the wider developer community reacted, aggregated from 5 discussions and 135 comments across lobsters, hackernews (as of 2026-08-12). **TL;DR:** Discussion sprawled far beyond the essay's data-science argument into a long, often heated debate about sex, gender, and trans identity, with only a handful of comments engaging directly with facial recognition, advertising data, and consent as concrete harms. **Sentiment:** 15% positive · 25% mixed · 60% skeptical **The case for** - Some pointed to concrete harms, like being unknowingly included in facial-recognition training data or ad networks inferring gender for profiling, as evidence the essay's concerns are real. - A few argued that removing unnecessary gender fields (e.g. using neutral titles instead of gender categories) would sidestep most of the harms discussed. - Several detailed, well-received explanations argued that biological sex itself resists clean binary categorization, undercutting the idea that 'objective' data classification is neutral or harmless. **The pushback** - Many pushed back that gender/sex data is sometimes practically necessary (advertising, security checks, medical dosing, sports) and that refusing to collect it isn't always feasible. - A large contingent disputed the essay's framing entirely, arguing gender identity claims lack the same evidentiary basis as biological sex and that doctors' gatekeeping is scientifically justified rather than bigoted. - Some felt the essay's swipe calling doctors and neuroscientists bigots-by-implication was an unnecessary, alienating overgeneralization that distracted from its actual argument. - A few argued the piece scapegoats 'data science' specifically when the deeper critique applies to categorization in science broadly, or to society rather than the tools themselves. **By community** - lobsters (heated): A single ambiguous sentence about doctors/bigots kicked off an extremely long, contentious detour into sex-vs-gender science and trans identity politics, with the essay's actual data-science thesis discussed only intermittently and mostly in the early and late parts of the thread. **Hottest debate:** Whether gender is a purely subjective, socially constructed category or something doctors can objectively test for based on biology, which spiraled into a much larger dispute about trans identity itself. **Open questions** - Is there any legitimate technical use case for automated gender/queerness detection that isn't advertising, and if so, does it justify the harms described? - How should systems that must record sex or gender (medical, security, legal ID) do so without enabling the surveillance harms the essay describes? **Highlights** > As a trans woman who recently discovered that my photos are being used without my consent to train facial recognition to identify trans women, I heartily suggest that cis and straight engineers read this and then consider whether tracking data points like gender are actually necessary. > — [inactive-user on lobsters · 41 points, 2 comments](https://lobste.rs/s/h9haqa/counting_countless_why_data_science_is#c_appyug) > > transgenderism is very similar to Catholic transubstantiation; a believe about the metaphysical nature of reality. Oh you are on SUCH a good track here. I'd like to take this point a little further. Gender *itself* is a metaphysical belief. So is sex. Both sex and gender, as equivalence classes that allow you to categorize all people as one of, say, two types, are broken. They're fuzzy categories at best, and once you start getting rigorous, you start seeing lots of exceptions. That's not to say they're not broadly useful concepts, and they do work for lots of people, but when we're designing systems that work with sex and gender (for instance, in medicine, or sports!), we should think carefully about what those categories *really* mean, how they'll be used, and how they're going to work for people who don't fit the boxes neatly. For instance, say you're trying to code people as male or female based on the "rational physical world" of measurements. Do you take chromosomes? Sure. 46XY is male, 46XX is female. I guess we could call 45X female and 45Y male too. What are 47XXX, 47XYY, and 47XXY? 49XXXXY? Y *generally* induces testicular development, so maybe any copy of Y codes as male. But if you've got androgen insensitivity syndrome, some "standard male" 46XY people will have breasts and labia. And some of those 46XX folks with CAH might look outwardly like men. Which soccer team to do you force these people to play on? Who gets prompted for testicular screenings and pap smears? We haven't even gotten into chimerism yet. You know some people have multiple genotypes, right? Okay. Let's back off to genitals. We can measure those, right? If you have testes, you're male. Unless you've had a hysterectomy, or orchiectomy, but you'd *know* if that happened to you. Well, unless it happened at birth. Turns out that in some cases a doctor *guesses*, based on some complicated rules and their best judgement, what kind of genitals would be best for you, a newborn infant. Maybe you're genetically XY, but your penis was small at birth, so they surgically constructed a vagina, removed the testes, slapped an F on the birth certificate, and told your parents you should be raised female. Maybe you never knew about that until getting a gene test in middle age, and then you realized, holy shit, I've been male/intersex/trans/etc this whole time. Maybe you've got ambiguous genitalia. Maybe you've got a vagina, but no uterus, and internal testes instead. These things aren't THAT rare. Roughly one in two hundred people don't have a "standard" male/female body plan. Chances are you personally know someone like this! About one in a thousand people receive "normalizing" surgery. If you're a primary care physician, you've likely got a couple patients who might need you to have a more nuanced understanding of what sex is. Maybe you need to ask your male patients if they're considering pregnancy. Fine. Let's decide based on someone's body's secondary sex characteristics. Adam's apples? Facial hair? Voice? Musculature? Height? Breasts? Pelvic geometry? These are all over the place. Slender, wispy people covered in hair. Barrel-chested, high-voiced musclebears with no Adam's apples. People who fall *smack dab in the middle* of your female and male coordinates on whatever measurement system you choose. Some of these characteristics are reshapable with surgery, hormones, and practice. Some change with age. But these secondary characteristics are also how we usually code people as male or female. For example, when was the last time you checked to see if everyone walking into the men's bathroom had a penis? And this is just sex! Physical, biological stuff that's measurable with rulers, blood tests, and CT scans. Gender's even more spectacular. > — [aphyr on lobsters · 19 points, 2 comments](https://lobste.rs/s/h9haqa/counting_countless_why_data_science_is#c_ca6s6l) > > 99.9% of all humans have normal chromosome Karyotype. What you describe are outlies of less than 1% Sure! XXY is ~1:1000 births. If we were doing research on something with broad interest, like a soda flavor, we might decide that the error we introduce by categorizing those people as male, or female, or removing them from the sample altogether, isn't significant enough to affect our results, and go on with our lives. It'd be problematic if we needed *everyone* in a sample of tens of thousands of people to fit exactly one category. Unfortunately, that's how a lot of our society is built. XXY doesn't shorten life expectancy much. That means in the US you've got ~327,000 people with XXY. A suburban high school with a population of 2000 students has a good chance of having at least one XXY person in their prospective student pool, and will need to decide whether those people are allowed admission (for an all-boys school), and what sports teams (if sex-segregated) they play on. A typical Wal-mart will see a handful a day, and expect them to choose the right bathroom. A suburb of 50,000 could (with sufficient motivation) field an all-XXY football team. A decent-sized airport like Miami International will see ~120 passengers *every day* whose sex indicated on their ID doesn't match their karyotype. A major metro area could run a Kleinfelter Kickball league. Let's imagine you're like me: a cis man who hasn't been karyotyped. Maybe you took a 23andme test, and that came back XY, because they don't report trisomies. There's a nontrivial chance that you, who grew up believing you were male, who has always been treated like a man, might not be a "biological male"--at least by a strictly genotypic definition. Of course that's ridiculous. You *know* you're a man, and everyone else does too. You don't get dirty looks for using the men's room at work. You can hand over your ID at the bar or airport without fear. Nobody called the school board outraged that you were allowed to play on the men's basketball team. You never had to worry about these things, and you still don't. That'd be silly. Nobody's going to find out unless you tell them, and even then, I'm sure they'd understand. Well, most people would. Except for people who believe that "male" means "XY", and then, well. That's a different story. That story might involve your marriage being challenged in court--but a neighboring state used anatomy, rather than genetics, to determine sex, so you were able to remarry there and bypass the issue. You may have been asked to use the women's room at work--only they don't want you there either, so you wind up having to ask for a special bathroom. The only single-occupancy stall is on another floor, so you take the elevator every time you have to pee, and hope nobody's using it. You might be stripped of your sports trophies, and banned from competition. You might wind up with a new ID that says F on it. You have to sigh and explain every time you show it, and hope that when you're pulled over it doesn't lead to arrest. You haven't gotten the bank updated yet, and you're dreading the bureaucratic nightmare that's sure to follow. Every few weeks, you blunder into a conversation on the internet where someone insists that you aren't a real man. Think about how that might feel. Now read your posts again. > — [aphyr on lobsters · 13 points, 1 comments](https://lobste.rs/s/h9haqa/counting_countless_why_data_science_is#c_k4m3hg) > I think perhaps you misunderstand - I specifically object to people developing software that identifies faces as "trans" or "not trans", not the inclusion of trans people in facial recognition datasets in general, and especially the lack of any way to remove my data from that dataset. Edit: Furthermore, this demonstrates a lack of understanding of the power of monopoly. For instance, I could not possibly care less if some jackass writes millimeter wave scanner software that consistently identifies me as a threat because my body doesn't conform to their ideas of what a body should look like (because I'm either a man with odd chest protrusions or a woman with an oddly shaped crotch.) It starts to matter a hell of a lot when that software decides whether or not I get patted down whenever I fly. If your software is going to be used in a situation where people have no choice but to interact with it, and there are consequences for it failing to classify them right, it is _your_ responsibility to make it work right, for _everyone_, or make it right to them. You don't get to shift that off onto them. > — [inactive-user on lobsters · 27 points, 1 comments](https://lobste.rs/s/h9haqa/counting_countless_why_data_science_is#c_kd6zrl) > > I’m still deeply upset by what they are doing. Like @NoraCodes said: What is the issue here? My queer activism is centered in things like: - I should be able to see my partner in the hospital - Everyone should have access to HIV prevention - People should be called by the names and pronouns they like - You should be able to get hormones, hair treatment, and surgery if it helps you live a fulfilling life - I'd like folks to stop yelling "faggot" at me on the street, and I want to buy groceries without people threatening to kill me for being gay. - It'd be cool if kids didn't grow up feeling like they were fundamentally unlovable and worthless because they were queer - Let's... not murder people for their sexuality or gender, ok? - You should be able to use the restroom without being assaulted because you didn't seem masc or femme enough to be there I hope these aren't too upsetting for you. > — [aphyr on lobsters · 22 points, 1 comments](https://lobste.rs/s/h9haqa/counting_countless_why_data_science_is#c_dpon3v) **Source threads** - [lobsters](https://lobste.rs/s/h9haqa/counting_countless_why_data_science_is) · 51 points · 133 comments - [hackernews](https://news.ycombinator.com/item?id=21265012) · 3 points · 2 comments - [hackernews](https://news.ycombinator.com/item?id=19618187) · 2 points · 0 comments - [hackernews](https://news.ycombinator.com/item?id=21255221) · 1 points · 0 comments - [hackernews](https://news.ycombinator.com/item?id=19779054) · 1 points · 0 comments ## Similar posts on daily.dev - [The World Slash](https://daily.dev/posts/the-world-slash-cxuwvu9da) · Medium · 0 upvotes · 0 comments - [Beyond Numbers: How to Humanize Your Data & Analysis](https://daily.dev/posts/beyond-numbers-how-to-humanize-your-data-analysis-rzxkbtuug) · Towards Data Science · 0 upvotes · 0 comments - [Banning noise will be a disaster for statistical data products](https://daily.dev/posts/banning-noise-will-be-a-disaster-for-statistical-data-products-e2ebziowf) · Lobsters · 0 upvotes · 0 comments - [Schneier on Security](https://daily.dev/posts/schneier-on-security-ea5va2swz) · Schneier on Security · 0 upvotes · 0 comments --- Tags: [#data-science](https://daily.dev/tags/data-science), [#privacy](https://daily.dev/tags/privacy), [#ethics](https://daily.dev/tags/ethics), [#facial-recognition](https://daily.dev/tags/facial-recognition) [View this post on daily.dev](https://daily.dev/posts/counting-the-countless-why-data-science-is-a-profound-threat-for-queer-people-bsfcnbrhb) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Counting the Countless: Why data science is a profound threat for queer people","url":"https://daily.dev/posts/counting-the-countless-why-data-science-is-a-profound-threat-for-queer-people-bsfcnbrhb","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/counting-the-countless-why-data-science-is-a-profound-threat-for-queer-people-bsfcnbrhb"},"datePublished":"2026-08-12T11:22:24.049Z","dateModified":"2026-08-12T11:23:12.248Z","description":"An essay arguing that data science, as currently practiced, is structurally hostile to trans and queer existence because it demands quantification,...","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/71921a5652541975152ecf504ad19cb4?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/71921a5652541975152ecf504ad19cb4?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Lobsters","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Lobsters","logo":"https://media.daily.dev/image/upload/s--tl8v_Fku--/f_auto,t_logo/v1698841318/logos/lobste.jpg","url":"https://daily.dev/sources/lobsters"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/counting-the-countless-why-data-science-is-a-profound-threat-for-queer-people-bsfcnbrhb","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"data-science,privacy,ethics,facial-recognition","timeRequired":"PT15M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Lobsters","item":"https://daily.dev/sources/lobsters"},{"@type":"ListItem","position":3,"name":"Counting the Countless: Why data science is a profound threat for queer people"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/counting-the-countless-why-data-science-is-a-profound-threat-for-queer-people-bsfcnbrhb#faq","mainEntity":[{"@type":"Question","name":"Why did Joy Buolamwini's Gender Shades research recommend more diverse datasets for facial recognition, and what's the critique of that approach?","acceptedAnswer":{"@type":"Answer","text":"Joy Buolamwini's 2018 intersectional analysis found facial-recognition and gender-recognition systems biased against dark-skinned, female-coded people, and her follow-up recommendation was to use more diverse training datasets. Critics argue this reformist fix ignores that these systems rest on an essentialist, biological view of gender that harms trans people regardless of accuracy, and that making facial recognition better at recognizing marginalized groups mainly sharpens law-enforcement surveillance tools already used against them. Anyone weighing fairness fixes against surveillance risk in ML systems can track this debate on daily.dev."}},{"@type":"Question","name":"What is Dean Spade's concept of administrative violence?","acceptedAnswer":{"@type":"Answer","text":"Administrative violence, coined by legal scholar Dean Spade, describes how administrative systems like government IDs and legal processes force people into narrow, rigid gender categories in order to access basic needs such as banking, healthcare, or a driver's license. Spade argues that reforming these systems, such as adding more gender options to IDs, tends to legitimize state control over gender rather than removing the underlying coercion, and mainly benefits the least marginalized. Developers building identity and data systems can follow this kind of ethics discussion on daily.dev."}}]} ```