tante

Sociotechnologist, writer and speaker working on tech and its social impact. Communist. Feminist. Antifascist. Luddite.

Last Notes

2025-03-28 09:18:38 UTC

Vulgar Display of Power
Hayao Miyasaki is the co-founder of Studio Ghibli, a Japanese animation studio known worldwide for their stunning, emotional, beautiful stories and movies. At the core of Studio Ghibli’s work is a deep engagement with questions of humanity. About what it means to be a human, about how to care for one another and the world around us. But also about the levels of cruelty that humanity can be capable of – and how there still are grace and love even under those conditions. Studio Ghibli’s work is distinct. In the level of quality over the years, in their very recognisable art style.
Hayao Miyasaki also is know for his reaction to a few technologists showing him something we’d call “AI” today to generate “creepy” moving figures:
https://www.youtube.com/watch?v=ngZ0K3lWKRc
His reaction to the proud developers showing an animation of a thing using its head to move forward claiming “Artificial intelligence could present us grotesque movements that we humans can’t imagine.” is very telling:“Every morning, not in recent days, I see my friend who has a disability. It’s so hard for him just to do a high five; his arm with stiff muscle can’t reach out to my hand. Now, thinking of him, I can’t watch this stuff and find it interesting. Whoever creates this stuff has no idea what pain is.
I am utterly disgusted. If you really want to make creepy stuff, you can go ahead and do it, but I would never wish to incorporate this technology into my work at all. I strongly feel that this is an insult to life itself.”
Hayao Miyazaki
He capped that off with a scathing remark about what had happened:“I feel like we are nearing the end of times. We humans are losing faith in ourselves.”
Hayao Miyazaki
Now of course he wasn’t looking at modern “AI” systems like the ones Microsoft/OpenAI/Anthropic/etc. are trying (with little economic success) to sell. But his argument would apply just the same: “Whoever creates this stuff has no idea what pain is.” Or joy. Or love. Or hunger. Or longing. Or need. Or want. All the parts of us that define our humanity and our way of interacting with the world. The things that make us care, that push us towards doing something, finding something out. Saying something and hearing something.
For the longest time OpenAI’s systems would try to block people from generating images in the style of certain artists. This was obviously for copyright reasons, the didn’t want to get sued (even more than they already are). Which is something they just changed very explicitly. You can now easily generate stuff in the style of Studio Ghibli and Sam Altman made his avatar on X-The Nazi Network a ghiblified version of himself.
And I think that on one level David Gerard is right framing it as a distraction from their money problems:This means OpenAI has to make more and more announcements so they look to the investors like they’re still cool and interesting. Any old garbage will do, like a literary fiction writer bot or something.
David Gerard
But I do think it does go further. There is a reason they chose Studio Ghibli. Sure, its style is very cute, very distinct, but that is not the whole story. It’s not that they just picked something cute and accidentally the co-founder of that studio hates their whole approach from the bottom of it’s hard. OpenAI picked Studio Ghibli because Miyazaki hates their approach.
It is a display of power: You as an artist, an animator, an illustrator, a writer, any creative person are powerless. We will take what we want and do what we want. Because we can.
Because we can. This is the idea of might makes right. The banner that every totalitarian and fascist government rallied under. OpenAI luckily is no government but their ideas, their thinking has influenced many current governments all over the planet. “If we are not allowed to take everything we want without payment and against people’s will, we will never create the machine god to solve all our problems.”
OpenAI’s move is an attempt to see what the reaction to them explicitly, willingly, gleefully breaching another boundary, acting against the explicit and known will of the people they use their machines on. And many in the public seem to eat it up, turning their holiday pics into “Ghibli-style” images.
The scholar of authoritarianism Timothy Snyder (who just left his position at Yale University going to Toronto in a very non-subtle reaction to growing Fascism in the US) write his book “On Tyranny” to give people 20 lessons about how to react to and resist tyrannical movements. The first rule is: “Do not obey in advance.“
This is what many are doing here. Submitting to this logic of injustice and domination in advance. Because it plays well on Instagram or because “it’s just pictures”. But this is submission to a logic of dominance by those who have power. It is a submission of democratic rights and understandings. It is a submission to a vulgar display of power.
And that is why the White House used that tech to [content warning about the following link: It’s a obscene display of violence] illustrate their cruelty against migrants. The cruelty is the point. But so are the other forms of violence displayed. Liked it? Take a second to support tante on Patreon!
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
#ai #dominance #fascism #ghibli #miyasaki #power
https://tante.cc/wp-content/uploads/2025/03/Mononoke_hime_cgi.png

2025-03-21 16:13:20 UTC

That asterisk …
So whenever you use some AI chatbot (regardless of the maker), it’s there. That motherfucking asterisk.
Like look:
Quote: “Gemini can make mistakes […] so double-check it.”
I have a question (well multiple ones but let’s start with my biggest one): Why do we accept that?
Say I order groceries from my supermarket. Some fruits, maybe some milk, yoghurt, whatever. Does the supermarket add that kind of asterisk? Do I get a card attached to my order saying “Yeah this might all be super spoiled or even the wrong stuff or laced with poison, better check it”. I don’t know about your supermarket but I haven’t seen that.
Now admittedly I rarely order groceries but still: Why did we come to accept a whole industry releasing products that they all have to affix an asterisk to basically telling us that their product doesn’t work?
No product is flawless, every manufacturer makes mistakes. It happens. Still we never see those asterisks, why? Because if something goes wrong the supplier has to fix things, reimburse you, maybe even pay you some fee. And not only when you sue them.
I don’t drink coffee, but I do very much like a type of caffeinated cocoa with peppermint (I know, it sounds weird, but it’s delicious!). After I ordered my last batch a few days later (I hadn’t opened the package yet) I got an email from the manufacturer that their current batch had some issues (mainly not enough peppermint in there) so they reimbursed me. The powder wasn’t even poisonous, it was just not the thing I paid for. And they fixed it (and added a coupon for my next purchase but that was just good service). This is how we are used to things working: If you release something to the market, you take responsibility that the stuff you are delivering is the stuff you said you would deliver, the stuff your customer paid for.
I find it really weird that – just because AI is magic and all – we allow huge corporations to release something that is so obviously not up to standard that they have to remind you about it with every use. How is that acceptable? Why do we accept that with AI companies and not anywhere else?
If your product needs that big of an asterisk to be put out there maybe you really don’t have a product but a cute demonstrator. That you possibly shouldn’t charge for and that nobody serious should integrate into processes they care about.Liked it? Take a second to support tante on Patreon!
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
#ai #comment
https://tante.cc/wp-content/uploads/2025/03/pexels-photo-6250312-1200x900.jpeg
https://tante.cc/wp-content/uploads/2025/03/grafik-1200x271.png

2025-03-16 12:59:12 UTC

It’s all hallucinations
The discourse on “AI” systems, chat bots, “assistants” and “research helpers” is defined by a lot of future promises. Those systems are disfunctional or at least not working great right now but there’s the promise of things getting better in the future.
Which is how we often perceive tech to work: Early versions might be a bit wonky, but there’s constant iteration and work going on to improve systems to be more capable, more robust and maybe even cheaper at some point.
The most pressing problem for many modern “AI” systems, especially the generative systems that are all the rage these days are so-called “hallucinations” which is a term describing when an AI system generates incorrect information. Think a research agent inventing a paper to quote from that doesn’t exist for example (Google’s AI assistant telling you to put glue on pizza is not a hallucination in that regard because that is just regurgitating information from Reddit that every toddler would recognize as a joke). Hallucinations are the big issue that many researchers are trying to address – which mixed results. Methods like RAG are shifting the probabilities a bit but are still not solving the problem: Hallucinations keep happening.
But I think that this discourse misses an important thing: Anything an LLM generates is a hallucination.
That doesn’t mean that everything LLMs generate is incorrect, far from it. What I am referencing is what hallucinations are actually defined as: A hallucination is a perception you have that is not connected to any actual stimulus. You hallucinate when you perceive something in the world that you have no sensor data for.
The term hallucination itself is an anthropomorphization of those statistical systems. They don’t “know”, or “think” or “lie” or do any such things. They iteratively calculate the most probable set of words and characters based on the original data. But if we look at how it is applied to “AI”s I think there is a big misunderstanding because it creates a difference between true and false statements that just isn’t there.
For humans we separate “real perceptions” from hallucinations by the link to sensor data/stimulants: If there is an actual stimulant of you feeling a touch it’s real, if you just think you are being touched, it’s a hallucination. But for LLMs that distinction is meaningless.
A line of text that is true has – for the LLM – absolutely no different quality than one that is false. There is no long to reality, no sensor data or anchoring, there’s just the data one was trained on (that also doesn’t necessarily have any connection to reality). If using the term hallucination is useful to describe LLM output it is to illustrate the quality of all output. Everything an LLM generates is a hallucination, some might accidentally be true.
And in that understanding the terminology might actually be enlightening, might actually help people understand what those systems are doing and where it might be appropriate to use and – more importantly – where not.Liked it? Take a second to support tante on Patreon!
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
#ai #hallucination
https://tante.cc/wp-content/uploads/2025/03/DP808068-1200x1502.jpg

2025-03-12 12:41:43 UTC

Wie rechts ist die KI-Ästhetik?
In der taz argumentiert Annekathrin Kohout, man dürfe “KI”-Kunst und Bildgeneratoren nicht nur mit rechtem Gedankengut assoziieren, es nicht “den Rechten” überlassen:
Wer KI-Bilder vorschnell als „rechts“ abtut, verkennt ihre eigentliche Sprengkraft. Sie sind Ausdruck einer kollektiven visuellen Imagination und kraftvolle „Wunschmaschinen“,
Source: Wie rechts ist die KI-Ästhetik?: Bildpolitiken des Affekts | taz.de
Ich glaube, dieser Beitrag verkennt ein wenig, warum diese Assoziation mit rechter Ideologie so stark und prägend ist und weshalb man hier eben nicht ein Werkzeug hat, welches auch für Linke ergiebig ist.
Wie Gareth Watkins in seinem vielbeachteten Essay “AI: The New Aesthetics of Fascism” schreibt:
The right loves AI-generated imagery. In a short time, a full half of the political spectrum has collectively fallen for the glossy, disturbing visuals created by generative AI. Despite its proponents having little love, or talent, for any form of artistic expression, right wing visual culture once ranged from memorable election-year posters to ‘terrorwave’. Today it is slop, almost totally. Why? To understand it, we must consider the right’s hatred of working people, its (more than) mutual embrace of the tech industry and, primarily, its profound rejection of Enlightenment humanism.
“KI”-Bildgeneratoren sind strukturell, fundamental eine Gegenbewegung gegen die kreative Autonomie von Menschen. Ihre gesellschaftliche Relevanz leitet sich einerseits ab aus Versprechen von hoher Produktivität (schneller Content erzeugen) wie auch in ihrer Entmachtung der Kreativen selbst: Nicht nur basieren diese Systeme auf unbegreifbaren Mengen von nicht lizensierten und oft gegen den Willen der ursprünglichen Künstler*innen akkumulierten Daten, die in einem Statistiksystem geronnen zur weiteren Ausbeutung durch kommerzielle Anbieter verwendet werden. Die Behauptungen von “Demokratisierung” reduzieren sich bei genauerer Analyse meist nur als mangelnde Wertschätzung der erarbeiteten Kompetenzen der Künstler*innen, als Abwertung derer, die ihr Leben damit “verschwenden”, Kompetenzen in Fähigkeiten auszubilden, die nicht in STEM Wissenschaften unterrichtet werden. Es ist die Behauptung, dass kreative Arbeit eh wertlos ist. Und damit natürlich explizit gegen Kreativarbeiter*innen ausgerichtet.
Kreative erzeugen aber ja nicht nur Grafiken für Werbespots oder LinkedIn Posts, die in 5 Zeilen eine komplette Psychoanalyse des/der Postenden erlauben: Sie sind ja oft zentrale Akteure in Prozessen der “visuellen Imagination”. Künstler schaffen eben Kunst. Und diese scheinbar triviale Aussage wird im ursprünglichen Text negiert.
Kunst ist mehr als reines Handwerk. Visuelle Künstler*innen lernen nicht nur den Umgang mit Werkzeugen und Materialien, sie entwickeln auch meist eine eigene Handschrift, einen eignen Zugang. Dieser ist immer auch beeinflusst durch andere Künstler*innen, aber wird trotzdem geprägt durch die eigene Geschichte, das eigene Erleben. Kunst die uns bewegt ist anders, und das muss gar nicht unbedingt nur visuell sein sondern kann sich auch im Framing der Arbeit finden oder in der Art wie sie sich politisch präsentiert.
Bildgeneratoren nehmen diese Prozesse der persönlichen Entwicklung und Bildung aus dem Spiel: Niemand muss mehr die Welt erleben und durch Bilder reflektieren, jede*r kann einfach eine durchschnittliche Mainstreamvisualisierung in dem üblichen KI-Generator Hyperpop Look machen. Doch gerade wenn es darum geht, über die Zukunft, über Visionen und Wünsche zu sprechen ist ja die Auseinandersetzung das, was zu interessanten Ergebnissen führt. Klar kann man sowas hinbehaupten für KI-Prozesse aber ist “Ich formuliere mein Prompt 13 mal um, damit ich ein Bild erzeuge, das ich nett finde” wirklich dasselbe wie die den Denkprozess permanent verändernde langsame, schrittweise, oft in Kommunikationsprozesse eingbettete Arbeit am Thema? Ohne in zu viel Nostaligie zu verfallen: Die Ergebnisse von “KI”-Bildgeneratoren erkennt man 10KM gegen den Wind, eben weil sie so aussehen wie sie aussehen. Durchschnittlich, immer mit dem Gefühl “das habe ich schon mal gesehen”.
Doch gerade wenn man über das Entwickeln von Zukunftsoptionen nachdenkt, reicht da “das hier ist die Hyperpop Version der durschnittlichsten vergangenen Dinge, die zu deinem Prompt schon mal gemacht wurden”? Die konservative, koloniale, weiße Vergangenheit, die die Trainingsdaten immer noch dominiert, immer wieder zu erbrechen erscheint mir eher nicht links zu sein. Welche politische Richtung hat denn die Rückkehr zu einer diffusen Vergangenheit als Kern ihrer Ideologie? Richtig, der Faschismus.
Bildgeneratoren sind – wie jede Technologie – nicht neutral. Und ihr Bias is klar in eine Richtung: Einerseits durch ihre technische Struktur, die auf Reproduktion von Vergangenheit ausgelegt ist und andererseits durch den politischen Kontext ihres Einsatzes, der Arbeiter*innen entmachten und sie damit ökonomisch und politisch kalt stellen will.
Für Linke ist in KI Bildgeneratoren aktueller Prägung sehr wenig zu holen, ich würde sogar so weit gehen, dass die Ablehnung und der Widerstand gegen KI Bildgeneratoren gelebter Antifaschismus ist. Oder um es mit Dan McQuillan (Autor von “Resisting AI“) zu sagen: “We come to bury ChatGPT, not to praise it.“Liked it? Take a second to support tante on Patreon!
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
#ai #comment #kunst #taz

2025-03-03 22:17:13 UTC

Who is Free Software for?
For a while I have been arguing that maybe there are some issues with the whole “Open*” movements, their founding myths and ideologies (see for example my talk at Fluconf). This criticism comes from a place of love. All the writing on this blog is licensed CC-BY-SA to allow others to take the texts and do something with it: I release my work under those conditions because I believe that we need strong and rich commons to flourish as a society but also as communities, groups and individuals. I’ve also been running my own personal systems (servers, my own laptops and a few other systems) on Linux for more than 20 years now. I am deeply embedded in the space of open culture but also Free and Open Source Software.
This morning I made a bit of an off-hand remark that summarized a few thoughts going on in my head:
Post by nostr:nprofile1qy2hwumn8ghj7un9d3shjtnddaehgu3wwp6kyqpqyhj4ua6580h3w0kucwwrtwl4uv9pg7m2drxzplq5x3364jczp36sfkq54uView on Mastodon
People within the Open* movements have done the impossible, have created whole encyclopedias, the most successful and most used kernel on the planet and a metric fuckton of custom, optimized operating systems, software libraries, and user facing programs. Have contributed to the commons to a degree that wouldn’t even have been credible within science fiction stories. Some of these systems – and I am not kidding here – should be considered the digital wonders of the world.
So why have we not “won”? Wikipedia might be considered to have won: By now it is the default digital source of information for large parts of the planet when trying to get to mere facts. But Wikipedia is an outlier in that regard.
Using Statcounter’s metrics Firefox – the Open Source browser that is not depending on what Google’s ad department wants to do – currently (March 2025) has a market share of 2.63%. And sure mobile platforms and Apple’s anti-competitive strategies when it comes to mobile have made it harder but even on desktops it’s just 6.3%. That’s absolutely not nothing, it’s millions of people. But not exactly something that shifts any sort of power and influence away from tech giants.
Again using Statcounter, Linux has a desktop market share of about 4%. Which is a lot given how hard the main competitors Microsoft and Apple are making the life of Linux developers and distribution builders. But still not even close to being an actually relevant player in the market for most purposes. Which everyone using Linux feels daily when asking for certain software vendors to release Linux versions of their stuff: “It’s not worth it.”
I got a great response to the Mastodon post I embedded above:Link: https://polymaths.social/@rl_dane/statuses/01JNEDT4T4F0PC137D45863N0S
First: R.L. Dane is completely right in their description about how “Open Source” is kind of the corporate version of “Free Software”. “Open Source” tried to strip the little politics that “Free Software” as a concept carried to make the whole scene more digestable to corporate entities (in any meaning of the word).
Second: I believe that they also nailed who Free Software is for. Quote:What we call #FOSS today was originally for hackers by hackers.
R.L. Dane
For hackers by hackers.
And I am afraid that we haven’t moved enough past that mindset.
That’s 100% not saying that nothing happened. A big deal has happened. Think of Outreachy that tries to get more people from diverse backgrounds internships to work on open source and build a career in software. Think of PyLadies who have put in so much work to give more women the opportunity and a safer path towards becoming active contributors into the Python ecosystem. And there are so many, many activities like that. So much work so many technical communities put into making the path towards becoming a member easier, more inclusive, fairer, etc. Those activities are fundamentally about fulfilling the Open Source promise: To give everyone the ability to have control over the software they use and the tool to build upon what’s already there. Mission … not accomplished but on its way, right?
But what are we doing? What are we trying to help “everyone” with?
We are trying to give more people the opportunity to “become hackers”. So they can profit off of all this stuff built by hackers for hackers. This isn’t a project to free all of us, it’s a project to give everyone a degree of freedom if they join our club. If they assimilate. This is Borg-mode.
We are not meeting people where they are. We expect them to come to us in order to understand why our values matter and are the best. Which – sorry to have to say so – they are not.
In 1971 the black civil rights activist Fannie Lou Hamer coined the phrase
Nobody’s Free Until Everybody’s Free.
And we have not taken that to heard. Because it does not say “Nobody’s Free until everyone is like us and then we have tools to create some freedoms”.
This is why so many outreach programs don’t work. Because what we are selling isn’t a solution to people’s actual problems but a new identity. And most people already got one of those. Now the solutions we propose might actually help our target audience with a real problem that they are faced with but so often our narratives don’t connect to their realities. We’re stuck in our own heads. Our own mechanisms and traditions.
I keep realizing and feeling this being stuck whenever I talk about moving beyond Open Source and Licenses and all that. I get a lot of responses arguing for example that “if you restrict people to not use your software for war you are no longer compatible with Freedom 0 and are no longer Free Software” as if that mattered. Yeah sure, that’s the legal regiment we’ve built. And where has that got us? Are we happy here? Is that enough?
Is me being able to customize my systems to my needs good enough? I recently changed a lot of my infrastructure to depend less on US companies and service providers for the simple reason that currently hosting stuff in the US (digitally but also physically) does not feel save. I can do that. Can my dad? My neighbors? Is that their fault?
In motherfucking 2013 I wrote:Telling people to “host your own” when some big company closes or buys a service is very similar to the princess who, when learning that the peasants had no bread, said: “Let them eat cake.”Hosting your own is a solution for the gifted and wealthy few, for many it’s blatant cynicism.
a younger tante in 2013
I stand by that. We need to get out of our comfort zones and modes of operation. Need to move beyond the seemingly apolitical cyberspace of free licenses. We need to reshape our thinking towards more political goals and values.
Maybe then non-hackers might also give a shit.Liked it? Take a second to support tante on Patreon!
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
#floss #foss #free #licenses #open #openSource
https://tante.cc/wp-content/uploads/2025/03/pexels-photo-4709289-1200x800.jpeg
https://tante.cc/wp-content/uploads/2025/03/Screenshot-From-2025-03-03-22-43-00.png

2025-02-06 23:12:25 UTC

The Open Source Torment Nexus
One of the most popular memes when it comes to talking about the tech sector is Alex Blechman’s tweet about the Torment Nexus:
One can interpret the tweet in a bunch of ways. As a comment on the level of reading comprehension that billionaire tech leaders like Mark Zuckerberg or Elon Musk or some ghoul that Peter Thiel funded constantly show: Calling one’s products or comapnies “Metaverse” and “Palantir” does not express an understanding of who and what is criticized in the respective works of fiction. And we are not even asking for nuance here.
Another interpretation is that the products that are being brought to market as massive innovations seem to be less than stellar. Best case scenario they are just weird stuff nobody asked for (think Juicero), most of the time they are “here’s a thing you kinda want but you have to pay rent for it now instead of buying it and we might shut it down any time”. And a lot of the time they are just bad for you, products that insult your existence as a human being.
Obviously there is no objective way to make these calls, to decide what is dumb but meaningless or what is evil. People have different values, different needs and sometimes can accept different levels of crap for a specific product or service.
We talked about this a bit in the Q&A part of my talk at Fluconf that gave a bit of criticism of the Open Source movement and its beliefs. But while my argument in the talk was mostly that the beliefs that Open Source is based on might not actually be based on or support the actual political values that people might have there is value to having Open Source software. Software that you can use and change to suit your needs and demands (if you can of course).
But Open Source software does not happen in a vacuum, it is written by people. Often to serve their needs, sometimes to serve a community’s needs: To provide a better solution for users. The “Open Source Alternative to proprietary X” thing.
If you’ve been on the Internet for about 7 seconds you will have come into contact with guys (using a gendered term here because they are mostly male) who will respond to any criticism of or issue you have with a proprietary software with “Just use this Open Source tool”. It’s a bit annoying because a lot of the time that answer doesn’t come from a genuine desire to help based on empathy and an understanding of the problem the original post expressed. But It’s also oftentimes better than one might think.
The Open Source community has gotten quite good at replacing proprietary tools with open source solutions. The feature set might not be 100% there and some things might not work but there are totally valid alternatives to Slack or Chrome or Microsoft Office (if you really just need office tools and not something that specifically depends on MS’s ideosyncracies). There are whole websites dedicated to listing Open Source alternatives to proprietary tools with at least one of those alternatives being pretty much a copy.
This has tremendous value and provides a bunch of people with the software tools to do their thing – me included. But when building those replacements the Q&A at FluConf made me wonder: Are we maybe doing a tech-bro? Are we not properly reading and understanding the texts we reference?
Any artifact is a text. You can read for example any piece of software as a document about the beliefs and assumptions of the programmer about the users. The software tells you how the programmer sees and understands the user. It’s just a form of text that is maybe a bit harder to read and it might have animations and sounds and can send bits somewhere.
What triggered this train of thought was me thinking about social media platforms. With Meta and X and all US tech companies finding their love for fascism currently many people want to jump off those platforms to find a better home. And there are Open Source Alternatives to Twitter/Instagram/WhatsApp/etc. Which in general is good. Maybe. Sometimes.
(Quick sidebar: I’m gonna use an example of an Open Source project here but don’t read this as me shitting on that project. People can build what they want and have their reasons for doing things. My feelings and needs are not “right” and other perspectives are wrong. I’m using the project as an illustration.)
I don’t use Instagram. Not because I don’t like seeing pictures or because it is “beneath me” and not even because it’s Meta and Mark Zuckerberg bend the knee to Trump. (In fact I have an account I use to follow tattoo artists but that I only use for exactly that research because tattooers live on Instagram sadly) I don’t use Instagram because it’s bad for my mental health. The affordances and the social practices of Instagram have lead to every picture (and the people and lives in it) looking like an ad. Perfectly designed and sculpted representations of better lives of better people. It just makes me feel bad about myself, my looks, my life. For me Instagram is the Torment Nexus.
Now that might just be a me-problem. But when I see an Open Source project like Pixelfed that basically attempts to provide users with a federated, open source drop-in replacement for Instagram I wonder: Are we just building an Open Source Torment Nexus? Because Pixelfed looks like Instagram. Has all the bones of Instagram. The same basic logic of the app (maybe the recommendation algorithms aren’t as aggressive but you can always patch that). There is nothing that would lead me to believe that – should Pixelfed get very popular – it would not also develop similar esthetics. Would also help people feel bad about themselves (if they are vulnerable to that kind of thinking).
And Pixelfed isn’t the only tool here. Bluesky is basically a carbon copy of Twitter a few years ago (not technology wise but in the way that interactions work, the way in which status is produced). Mastodon isn’t hat different either. Which in this case doesn’t feel as bad for me – but I also liked Twitter.
It feels like too often we are bound to think about what’s possible in the framing of what businesses think is possible. I recently argued against scale and that topic is a perfect example: Meta needs any new product to scale to millions if not billions of users. Amazon needs to scale products to the whole world. Google needs to scale. Microsoft needs to scale. And they all need to do the things you need to do to scale. Build interaction loops that enforce you coming back to an app or platform regularly ideally daily, to make it a habit. Build tools promising you a view on everything, to give you feelings of power, understanding and control.
We here on the open, decentralized, free web don’t need to do that. Pixelfed does not need to scale. Mastodon does not need to scale. At least not in the way that Facebook needs to.
We are doing a tech-bro. Just like Peter Thiel calls his surveillance company after the stones the evil wizards use in The Lord of the Rings because he doesn’t seem to have understood that those are the baddies we are building tools that are built for growth hacks and scamming VCs out of their money because we don’t read them properly. Because we just take what’s there and make it open source.
Just is doing a lot of work in that last paragraph. The amount of work that goes into duplicating those proprietary functionalities is incredible, awe inspiring. Thousands of people put work into building those tools, often without payment, without a lot of thanks. Spending their limited time on this planet contributing to the commons with software.
But recognizing the amount of work and time and life that goes into building this huge pile of software shouldn’t we spend it on something good? Something that is actually good for the people using it? And are we always so sure that the things we build do that, enable that?
We should stop building the Open Source Torment Nexus. Because the problem with the Torment Nexus is not the software license or opaqueness of the code: It’s the part with the torment.Liked it? Take a second to support tante on Patreon!
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
#fluconf #openSource #tormentNexus
https://tante.cc/wp-content/uploads/2025/02/6sbhb1kcn7k-1200x800.jpg
https://tante.cc/wp-content/uploads/2025/02/fd2.jpg

2025-02-06 14:33:37 UTC

“A Luddite Criticism of Open Source” at FluConf
On February 1st I had the privilege of getting to speak at the first Fluconf about Open Source and why that whole movement might not do as much good for as as we might want it to. It’s not that Open Source is “bad” but that it is sometimes presented as a solution to sociopolitical issues it is not capable (nor even trying) to fix.
The talk description was:
“People interested in Fluconf will also be interested in Open Source or to be more precise Free/Libre Open Source Software. Everyone has been using Open Source for many years but many from this community have shifted to trying to run their personal infrastructures on non-proprietary pieces of software and some even hardware.
This is already an important shift towards freeing ourselves and each other from corporate dominance but is it really doing enough? Are our licenses protecting the values we actually care about?
Coming from a luddite background I want to dive into a a bit of a critical reading of the values that we use codified mostly as licenses: Are they really enough? Which aspects are they missing and why? And what are the consequences of those omissions?
I’ll try to end up looking at mechanisms of integrating luddist principles in software projects. Can they offer additional safeguards?
This sessions doesn’t claim to have all the answers. But hopefully some good questions and a few ideas of where to go.”
I embedded a copy of the video on Youtube but you can also watch it on Archive.org for less tracking and everything.
https://www.youtube.com/watch?v=_Ew_MlA5HbALiked it? Take a second to support tante on Patreon!
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
#fluconf #luddism #openSource
https://tante.cc/wp-content/uploads/2025/02/Bildschirmfoto-2025-02-06-um-13.41.11-1200x652.png

2024-04-22 19:03:50 UTC

“AI” as support systems (for diagnostics)
One of the more reasonable use cases for modern “AI” (statistical pattern matching and generating machines) is to support doctors in diagnostics, especially in the evaluation of complex data sets / documents in order to determine potentially dangerous abnormalities.
It’s a problem tailor-made for neural networks (it’s actually one of the types of cases with that we got trained on them back when I was still studying in the beginning of the 2000s): You have large more or less clean and tagged sets of data of diagnosed x-rays and such that you can feed into your system to train a detector. Now since human beings are not really standardized/normalized the results are never 100%, one person’s abnormal is another person’s normal. But you can get decent results.
Results that “AI” hypesters often celebrate: Look AI’s not a hype, studies show that AI systems are better than X% of doctors at diagnosing Y!
I think that many people don’t fully understand the context though: What we see time and time again is that these systems are better than a significant percentage of doctors at diagnoses. Cool. But which doctors?
More often than not you will find that these systems are better than new or unexperienced doctors but worse than experienced specialists. Which isn’t surprising. It’s like with text: ChatGPT and others generate text that’s decent-ish and better than what first semester students would write but usually a lot worse than what experienced professionals create. AI is cheap and mediocre.
Now we could say that that’s a positive: We’ve basically shifted the baseline up. No longer do you need to hope that a new doctor diagnoses your condition correctly, you have the AI level as baseline and more complex cases get sent to the experts (how many cases would be found as complex when the mediocre machine has already diagnosed the case is another very good question).
But there is an issue (one that you can see coming from a mile if you’ve every heard one of my talks on “AI”): How do young, inexperienced doctors learn, if the machines diagnose all the routine cases? You learn diagnoses by diagnosing, by talking about them with your peers, especially more experienced peers. It’s great that we’ve potentially made the baseline better but how do we ensure that young physicians get enough time to practice their diagnostic skills? Because the super experienced people will die at some point. And experience does neither come cheap nor quick.
Especially in a world where the main promise of “AI” is efficiency and productivity do we really believe that the “time savings” will be put into training young people who can’t yet do the work? When it takes years to get them to where they need to be to catch the stuff that the AIs didn’t find? In a different system we might but in neoliberal capitalism? In austerity?
It’s great that we look at reasonable use cases for statistics, especially in critical and highly complex domains like medicine. But the studies presenting successes need to be read and understood correctly and fully.
Better than X% of humans is an empty statement. You need to know what exactly characterizes the X% to understand the actual capabilities and limitations of the systems that are supposedly so great.Liked it? Take a second to support tante on Patreon!
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
#ai #diagnostics #medicine
https://tante.cc/wp-content/uploads/2015/07/IMG_20150717_164206-1200x599.jpg