Skip Navigation

  • Unfortunately many of the political issues with AI are left aside in this essay: the colonization of land for resources, for datacenters, the sociological aspects of a machine that trades truth for positive emotions, but one that is most important in an anarchism perspective is the active exploitation of non-white people for the benefit of white people:

    https://www.404media.co/ai-is-african-intelligence-the-workers-who-train-ai-are-fighting-back/

    AI, structurally, cannot exist without domination. It is impossible to have this technology work without a class exploiting another one.

    AI is at its core a political project to anthropomorphize a machine, give everyone the illusion it is autonomous and independent so that it can take decisions for you, but when things are bad the machine is the one to blame, so you have to treat it with respect. This pseudo-human isn't independent of course, it is completely under the control of other humans. The domination is the project.

    Whether those humans are fascists or democratically chosen doesn't change: at its core, AI is humans domination over another human form

    Maybe LLMs are useful, in another form. Machine learning has been a damn useful machinery with practical applications. But AI is a political project, and this project is the exact opposite of self-determination

  • In France violence is legally defined as harming people. Burning a Tesla shop is technically not violence, so it's perfectly fine 🔥

    Non-violence at best accomplishes one thing: make injustice public, force society to talk about it and position itself on a topic. It won't be enough to convice, unless you are a victim of publicized harsh violence (think blood is spellt, people get into a coma, or worse). Like all strategies revolutionaries will have, none should be discarded and they are all complementary. What matters is that everyone acts collectively and pushes in the same direction

  • While you are technically correct, you're implying that the "natural" state is a good enough state and nothing should be done about it.

    My house has walls and a door; it doesn't mean anyone can do anything they want with this. Even if the windows are clear, you're not supposed to install a camera that watches my bedroom. Even if the door is open, you're not supposed to open. A a society it has been decided that we should respect each other, respect each other's privacy. We have created rules, some written down and some implicit, for how to interact with each other.

    That is the point of OP. The "natural" state of whatever exists with the technical means, but that doesn't mean it's ok (or not ok): do we want to respect each other ? To take care of each other ? I very much want that, because the technical means should be only a means to an end, and in that end I want respect. The technical means, to me, must adapt to the end, not the other way around.

  • You didn't talk about self-hosting in your initial comment. S3 is not made for you, I guess it' ok to just move on to the next article

  • S3 is made for scaling. At an individual scale it doesn't matter. At scale, S3 moves maintenance time away from managing individual issues about something not working or slow here and there towards configuring and maintaining a consistent piece of architecture

  • No, it's made for entities managing data for thousands/millions of people. “overengineering" implies a specific scope, it's not the one you think about

  • atproto has no problem scaling. It's Bluesky, a platform that wants to see all the network, that has issues scaling

  • S3 is not made for you. It's made for entities that need to store millions of objects, with thousands of different rules, reading/writing from hundreds of machines without coordination, and with consistent, low latency. Now that some software use that as a storage layer, having an implementation for you is useful

  • Considering how much of energy is used for heating, this might still be interesting. But probably not for electricity generation

  • if a craptp is useful it's not crap anymore heh

    • Get yourself an account on neocities
    • Pick a website you like on https://512kb.club/ . Anyone of them, it doesn't have to be perfect
    • open the console to see its html code
    • use the inspector to pick elements (the title, a list, a link)
    • Read the code, and write it on your neocities site. It's important that you write it, not just copy-paste it, to really grasp the bits
    • Repeat with more and moro complex sites
  • Note that the people who made friendica also made hubzilla and (streams), and the latter is more or less the successor: ActivityPub only instead of other, project-specific protocols, continued development, ...

  • Not a chance unfortunately, bluesky doesn't work without an all-seeing-eye that brings people the content they want. At best there will be some satellites running on atproto, fully decentralized, but the core will still be bluesky and it will still be completely centralized

  • C'est de la mIArde @tarte.nuage-libre.fr

    Désactiver l'entrainement de l'IA de Github sur les infos du profil

    framapiaf.org /@booteille/116294666721710247
  • C'est de la mIArde @tarte.nuage-libre.fr

    Comment bloquer l'utilisateur "claude" dans github

  • I can only comment on tooling: fedify is very good, easy to get in and powerful enough for any usecase

    https://fedify.dev/

  • ai;dr

  • Heh, it's possible to integrate piefed inside zulip thanks to integrations. It's also certainly possible to integrate zulip inside piefed but that's another thing.

    Integrating zulip inside piefed is certainly interesting, maybe just an adapter of the piefed api over zulip's api should be enough. I don't see why the software should be checked though.

    Right now to me it's more important to improve piefed so that I can use its api without using too much resources, and with a bit more reactivity

  • hmmm that's interesting !

  • ai;dr

  • C'est de la mIArde @tarte.nuage-libre.fr

    Starlight Network No-AI List

    noai.starlightnet.work /list.html
  • Test PieFed @tarte.nuage-libre.fr

    encore un nouveau test

  • Test PieFed @tarte.nuage-libre.fr

    test nouveau post

  • C'est de la mIArde @tarte.nuage-libre.fr

    Une liste de projets libre et open source qui utilisent l'ia

    codeberg.org /ai-alternatives/llm-afflicted-software
  • PieFed help @piefed.social

    Unable to post a link with a tilde

  • C'est de la mIArde @tarte.nuage-libre.fr

    L’objection de conscience face à l’IA Générative est un moyen, pas une fin

    synthmedia.fr /edito/tribunes/lobjection-de-conscience-face-a-lia-generative-est-un-moyen-pas-une-fin/
  • C'est de la mIArde @tarte.nuage-libre.fr

    L'IA est un problème, et pas celui que vous croyez

    mail.paulineharmange.fr /archive/ia-probleme/
  • C'est de la mIArde @tarte.nuage-libre.fr

    Sticker contre l’IA générative – Pauline Harmange

    paulineharmange.fr /2026/01/sticker-contre-lia-generative/
  • C'est de la mIArde @tarte.nuage-libre.fr
    Featured

    Ces instances francophones qui refusent l'IA

  • New Communities @lemmy.world

    C'est de la mIArde

    tarte.nuage-libre.fr /c/cest_de_la_miarde
  • Fediverse @lemmy.world

    Do we need more users ?